US2832682A - Process for manufacturing special iron - Google Patents

Description

April 29, 1958 L. M. REYGAGNE PROCESS FOR MANUFACTURING SPECIAL IRON Filed June 18, 1953 2 Sheets-Sheet 1 United States Patent PROCESS FQR MANUFACTURENG SPECIAL IRDN Lonce M. Reygagne, Decazeville, France, assignor to Societe Metallurgique dlmphy, Paris, France, a French company Application June 18, 1953, Serial N0. 362,435

6 Claims. 01. 7s-41 The present application is a continuation in part of my prior application Ser. No. 111,905, filed on August 23, 1949, now abandoned, and concerns a process for the production of iron inside a shaft furnace in which the heat required is produced by the combustion of fuel or by electric power (blast furnace and low shaft furnaces operating on coal, or coke, or electric power, etc.), the sulphur content of the iron being always less than 0.04% and generally less than 0.02% and the silicon content being always less than 1% and ranging generally between 0.2 and 0.6%; said iron having a relatively high carbon content ranging generally from about 4% to 5.5%, and generally to about 5% and containing possibly manganese and also, in addition to the usual impurities, one or more additional elements as contained in materials of the burden, as it is well known, such as aluminium, nickel, vanadium, molybdenum and the like, employing highly basic and highly reducing slags which have a high alumina content consisting chiefly of dicalcium silicate and calcium aluminates which are soluble in water or in alkaline solutions and which may be possibly used for the manufacture of alumina.

The direct production of iron having low contents both of silicon and of sulphur, such iron being of particular interest in the manufacture of steel in open hearth furnaces, has been advantageously resolved by my invention.

The high carbon content of the iron obtained according to my invention is furthermore a benefit for various applications in foundries, for instance for casting ingot molds, rolls for rolling mills and the like.

It is hardly necessary to dwell on the technical ad vantages of very low sulphur content in such an iron.

As regards the silicon content of the iron, it is a fact long known that it is of interest in the production of steel in open hearth furnaces to lower the silicon content of the iron to the greatest possible extent, because the presence of this element makes the operation longer and requires larger amounts of basic fluxes, which increases the volume of the slag, raises the cost price and slows down the output. Experiments made on a large industrial scale have shown that when the silicon content of the iron charged into an open hearth ranges between 0.3 and 0.6% instead of between 0.8 and 1.1%, this allows lowering the additional amount of lime from 6 to 3.5% of the weight of the metal charge made up with 35 to 40% of liquid iron and 60% to 65% of scrap and this permits also shortening the operation by 10 to 11% and even by if the silicon content of the iron drops below 0.3%. It has been found moreover that it was possible in this way to increase the amount of the liquid iron and to substantially lower the losses of manganese and iron, which is particularly advantageous in those countries which are lacking in scrap and in manganese ore. It has also been found that the refractory lining is less subject to wear under such conditions.

Many processes have already been proposed for lowering the silicon content of iron. Thus it has beensuggested 2,832,682 Patented Apr. 29, 1958 ice on the one hand to mix the iron with slag from open hearth furnaces and, on the otherhand, to make the iron react with roll scale or with dust from blastv furnaces inside the casting channel or inside a ladle. Such treatments, performed outside the actual producing apparatus, are not only costly, but lead to a large loss of valuable manganese, the major portion of which goes in the slag and is lost for the steel making operation. It has also been proposed to remove thesilicon from the iron by making oxygen-containing gases bubble through the molten iron.

Further tests have also been carried out without much success for the direct production, in the blast furnace of.

iron from which the sulphur had been suitably removed and having a low siliconcontent, through the use of slag with very high lime content and the composition of which is similar to that of Portland cement, or else of highly aluminous slag of the type used for molten cement and having very low silicon content, say less than 8 to 10% while the lime content is comparatively low and at the utmost equal to that contained in monocalcium aluminate, or again by resorting to slags with a high content of magnesia. This proposed use of slags of such Portland cement or molten cement composition with very high lime or alumina contents, of very high melting temperature or being extremely viscous and which even show thesedrawe backs associated, leads to the necessity of operating the furnaces with very high hearth temperatures, that rise generally above 1600 C., with all the inherent risks and drawbacks resulting therefrom: high fuel consumption, poor behaviour of the refractory linings, breaking of the crucible and chiefly too important a reduction of the oxides other than iron oxide contained in the burden and chiefly of the silica, which produces an iron whose silicon content is actually higher than normal, in contradistincr tion with the aim of the operation. As to the use of slag having a high content of magnesia, it may lower the working temperature and the silicon content of the iron down to about 0.50%, but without thereby lowering the sulphur content as compared with the normal operation of the blast furnace.

It is therefore apparent that it has not been possible hitherto to solve in a truly economical and commercial way the problem of the production of iron that is low in silicon and very low in sulphur and in particular through direct production inside a shaft furnace'at working temperatures that are normal for such shaft furnaces.

Now, I have found a practical method for obtaining directly such an iron in all types of furnaces suitable for the reduction and melting of the ore and more particularly in shaft furnaces, this being produced at temperatures of operation of the furnaces and in particular of blast furnaces that are much lower than those that were considered hitherto as necessary and that are even lower than or at the utmost equal to the usual operating temperatures applied in such types of furnaces for the production of standard grade iron. With the method according to my invention, the amounts of slag produced are normal and range according to the raw material used between 0.3 ton and 1.5 ton and are generally less than 1 ton per ton of iron.

The process of the invention can be conveniently applied in numerous cases; in particular, the use shouldbe cited, for the manufacture of cast iron, of the so-called lateritic iron ores with aluminous gangue and more specially of those which contain at least one part alumina for one part silica. These iron ores, considerable fields of which exist in equatorial countries, are of great interest for iron and steel industries in America and Europe.

It is known that these ores are too aluminous, even when considering the other elements of the burden and erally the following composition:

. Percent Silica 30 to 36 Alumina 22 to 30 Lime 31 to 34 It also contains an important magnesia content to 8%) used as fluxing agent. This acid process yields irons. witha sulphur content which may sometimes be as high as 0.5%; the silicon content is relatively high. Because ,of the much too high content of sulphur, it is necessary to carry out desulphurization, for instance by using sodium carbonate, which entails technical as well as economic drawbacks. i

In the process of the invention, it is possible, as illustrated by the examples hereafter, to use ores having aluminous gangues, in a much more economical way, either alone, or with an addition of slight amounts of aluminous and calcareous fluxing agents, while keeping coke consumption and slag production, per ton of iron, within the usual limits of common raw materials, and usually lower than those obtained in the'acid process with the same aluminous gangue iron ore. Moreover, the slag produced is of greater value for it may be used for the recovery of alumina. I e

The following description and corresponding appended drawings show how it is possible to carry out my invention, the features appearing in the specification and in the drawings forming part of the invention as defined in the accompanying claims. In said drawings:

Fig. 1 is a portion of Rankins triangular diagram on which are illustrated the areas in which should lie the points illustrating the composition of the slags produced, said composition being given with reference to their three main components CaO, SiO A1 0 Fig. 2 is a diagram showing in ordinates thedifierence in temperatures between the temperatures T (R) given by Rankins diagram and the industrial melting temperatures t (I. M.) of the slags obtained according to the present invention, and in abscissae the temperature given by Rankins diagram; said industrial melting temperatures being defined hereinafter.

During my investigations concerning more particularly the production of iron in blast furnaces, I have found that the object sought for can be reached by the use of slags having a very high basic index CaO+MgO SiOg that is higher than 2.18, said slags containing at least two molecules of lime per molecule of silica, and more than one molecule of lime per molecule of alumina, and being generally richer in alumina than in silica, which slags allow working at the desired low operative temperature.

I have also found that such slags that are constituted substantially by dicalcium silicate and calcium aluminates containing one to three molecules of lime per molecule of alumina and as little as possible ternary combinations such as gehlenite, have a large desulphurising capacity and, contrarily to the usual belief, a high fluidity at the usual operating temperatures of blast furnaces, even when their silica content rises above 8 to 10%.

The general belief was, as a matter of fact, that it was dangerous and even impossible, by reason of the high melting temperature and the viscous nature of the slags obtained with silica contents higher than 8 or 10% to (i. e. the weight ratio use burdens that contain so much alumina and so much lime, and it was also believed that the only possible operative area was defined in the triangular silica-aluminalime diagram, generally termed Rankins diagram (as disclosed in the papers of Rankin, G. A., and Wright, F. E., Ternary System CaO, A1 0 SiO American Journal of Sciences, Series 4, vol. 39, 1915), by the figurative points located near or underneath the lines separating the area of dicalcium silicate from the calcium aluminates areas. This was believed to be true, because the diagram shows, in the vicinity of these lines, temperatures of beginning solidification, when equilibrium is reached, that are low enough to be consistent with the normal operating temperatures in blast furnaces. In practice, such very low silica contents being generally not consistent with the incorporation in the burden of the usual iron ore and coke,

said slag area was quite ignored in iron and steel manufacture.

As a result of my investigations, I have found and this is one of the main features of my invention, that said industrial slags containing silica, alumina and lime, whose silica content may even be higher than 8 to 10%, and possibly as high as substantially 22%, are sufliciently fluid to be handled at temperatures that are substantially lower than those given by Rankins diagram for figurative points corresponding to the actual ternary composition of these slags, as concerns silica, alumina and lime.

These facts appear to be explainable by the following reasons:

(a) In an industrial furnace, the procedure of forma tion of the slag is quite different from that obtained when operating in a laboratory through admixture of powdery silica, alumina and lime; it is possible to obtain, for similar elementary compositions, slags the crystallographic structure of which varies widely according to the more or less complete chemical combinations that have been achieved. Rankins ternary diagram of silica, alumina and lime contents, the portion of which corresponding to the slags that I use is illustrated in a simplified manner in Fig. l, presupposes as a matter of fact that thermodynamic equilibrium is completely obtained. However, this is never quite true in industrial practice because the materials do not remain suflicient time in the used metallurgic apparatuses for'permitting the completion of the reactions taking place therein, and some superfusion or supersaturation generally takes place. A slag, the temperature of which is comprised between that of the liq- .uidus and that of the solidus, and that is normally constituted by a mixture of the liquid and solid phases may even, with a large proportion of the latter, show a satisfactory fluidity and can be handled for industrial purposes in the same manner as a liquid. I have checked this by experimenting on synthetic mixture containing solely the three main components SiO A1 0 and CaO and I have found that, for ternary compositions to be used in accordance with my invention as defined hereinafter, there exists between the temperature at which said slags are sufficiently liquid and are capable of flowing and the temperature of beginning solidification according to the diagram, a diiference that may range between and-400 C. and that is generally of a magnitude comprised between and 250 C. In the present specification, I will call industrial melting temperature designated as t (I. M.) the temperature, below that indicated by Rankins diagram, at which a slag is in actual practice still fluid enough for it to be considered as a liquid capable of flowing easily. The diagram of accompanying Fig. 2 (in which the hatched area contains the experimental points) summarizes the results of these experiments; it shows that in the area of'the compositions examined, and for temperatures of beginning solidification,

. according to Rankins diagram, included between 1400 and 1800 C., said difierence between the two temperatures increases according to: a substantially linear func tion of said Rankins temperature of beginning solidificature), and that said difference, for the below referred to silica-alumina-lime slags, is 100 to 400 C., more generally 150 to 250.

(b) Furthermore, as the slags actually used in industry contain not only silica, alumina and lime but also other compounds such as iron, manganese and titanium oxides and also calcium sulphide and the like, their melting temperature is thereby lowered, the difference between the temperature of beginning solidification as given by Rankins diagram of the corresponding ternary silicaalumina-lime slag (the other components of which are not taken into account) and the industrial melting temperature of said industrial slag is still greater by 150 to 200 C. This temperature lowering effect is additional to that mentioned in the immediately preceding par-agraph and hence both these effects are additive.

Finally, in the case of an industrial slag corresponding to a ternary silica-alumina-lime composition illustrated by a figurative point on Rankins diagram located inside the dicalcium silicate area between the isotherm lines 1360 C. and 1800 C. (and somewhat higher) of said diagram, the temperature at which said slag is practically still liquid and may be handled as a liquid, i. e. its industrial melting temperature, is located much lower and substantially between 1210 and 1550 C.

Said facts have been checked on an industrial scale in the blast turn-ace by myself and I have found that it is possible to obtain a favourable fluidity and flow-ing capacity allowing normal operation of the blast furnace, with a hearth temperature measured at the cinder-notch that is less than 1550 0., inside an area of slag compositions containing more than 8% and up to 22% of silica, provided the alumina content is of at least 19% and the lime content higher than 48%, the figurative point corresponding thereto in the triangular diagram being located inside the dicalcium silicate area. 7 In accordance with my invention, the nature and the amount of raw materials used are chosen in such a manner that the slags have silica, alumina and lime contents defined by a point located inside an area MNOPQR, within the arear of dicalcium silicate in the triangular diagram, said area MNOPQR having as boundaries the straight lines corresponding to the following contents: SiO;,, 8%; SiO 22%; CaO, 59%; and 'by the curve separating the area of dicalcium silicate, on the one hand, from the areas of gehlenite (N), of monocalcium aluminate (OP), of tricalcium aluminate (part of QR) and of tricalcium silicate (remaining part of QR) on the other hand, part of the lime being replaced if desired by a molecularly equivalent amount of magnesia.

In other Words, the figurative point of the ternary slag corresponding to the industrial slag to be actually used is located in the portion of the area of dicalcium silicate that is defined by the straight lines: silica content 8%- silica content 22%lime content 59%.

Besides, it is important to note that according to the invention the slag corresponding to those figurative points which are characterised by a low melting point and a low viscosity, does not contain more than 63% of dicalcium silicate, or better has a dicalcium silicate content comprised between 23 and 63%, the main other compounds of such slag being pentacalcium trialuminate and monocalcium aluminate, with a preferable predominance of pentacalcium trialuminate.

In fact, I found that in contradistinction to the dicalcium silicate which, owing to its high melting point (about 2100 C.), causes a high melting point and a high viscosity of the slag, the pentacalcium trialuminate acts; as a flux, even in relatively small quantity, because of its lower melting temperature (about 1450" C.) as compared to that of the other calcium aluminates (monocalcium aluminate about 1600 C., tricalcium aluminate about 1650 (3.), because of the well-known eutectics law and of its other properties, speeding and increasing thus the fluidizing of the slag.

Owing to the above reasons, I resort preferably to a slag corresponding to a figurative point located inside the above referred to area MNOPQR, in a preferred smaller area mnpqr defined by lines corresponding to the following content:

SiO 10% SiO 20% C210, 49.5% A1 0 23% in which slag the dicalcium silicate content varies be tween 28.7% and only 57.3%. I

Owing to the low melting point of slags which are richer in pentacalcium trialuminate and with a lower content of dicalcium silicate, I better still resort preferably to a slag containing more than 23% pentacalcium trialuminate and less than 52% dicalcium silicate, and correspending to a representative point located in the ternary diagram in the smaller area ABCDEF bounded by the straight lines corresponding to the following content:

\ Percent CaO 50 SiO 11 S10 18 A1 0 25 and by two further straight lines joining respectively the figurative point SiO ZCaO (corresponding to SiO =34.8% and Ca0=65.2%) with the figurative points A1203, 1.3Ca0 (corresponding to Al O =58.29 and 0210:4111) and A1 0 1.9Ca0 (corresponding to Al O =49.04 and CaO=50.96).

It is a known fact (see for example Seailles U. S. application Serial No. 341,900, filed March 12, 1953) that slag wherein the alumina is combined so as to form aluminate, which is the case for the above defined slags, allows a ready recovery of the alumina contained therein, which forms a supplementary advantage of my improved method.

The theoretical temperature of beginning solidification, according to Rankins diagram, of slags having the above composition varies in a continuous and gradual relationship to their chemical composition, which is favourable for a smooth and reliable operation of the furnace. This theoretical temperature varies gradually from about 1360" C; to 1840" C. when the composition of the slag varies gradually from one end of the area considered to the other, the corresponding extreme compositions being substantially as follows:

For 1360 c; Percent SiU 8 A1 0 4s CaO 49 And for 1840 C:

SiO 22 A1 0 19 CaO ..59

Actual industrial slags corresponding to said area benefit also by this interesting property and their industrial melting temperature varies as disclosed hereinabove between 1210 and 1560 C., which is quite conducive to a smooth and reliable operation of the furnace. This fact makes the adjustment of the operationof the furnace, and chiefly that of its temperature for the purpose producing iron having the desired silicon contents withinthe above-disclosed range comprised between 0.2% and 1%, an easy matter.

".I'he steadiness of the silicon content in the iron, which is absolutely necessary for an adequate running of the steel furnace refining this iron is achieved without difiiculty, for the following reasons: (1) The silicon content of iron corresponds, for a given temperature, to a state of equilibrium between the silicon content of the iron and the silica content of the slag. In industry, the unintentional and. unavoidable variations in composition of the various substances forming the burden inevitably result in composition changes of the slag and the iron; the greater the silica content of slag, the greater the variation, in absolute value, of the silicon content of the iron, for a same variation percentage in silica of the slag. The slags used according to my invention have a low content of silica; this resulting practically in a much higher stability of the silicon content of the iron produced. (2) The adjustment of the silicon content is achieved by deliberately varying the silica content of the slag fend/or by altering the operating temperature of the hearth. With the slag composition used according to my invention, the melting temperature of slag, on which depends the operating temperature of the hearth, in creases as the silica content increases. The two actions are conjugate and the adjustment of the silicon content is thus advantageously facilitated.

Furthermore, I have found that the usual factors such as the composition of the charges, the proportion of coke,

the temperature and the oxygen content of the blast, the

composition of the slag, are not the only ones that may be resorted to for adjusting the temperature of the blast furnace and consequently the composition of the iron; this temperature does not depend only in fact on the melting temperature of the slag but also and chiefly on its formation temperature. I have found that if it is desired to obtain, within the range of compositions disclosed hereinabove, an iron of a predetermined composition, it is necessary to act on the formation temperature of the slag through a modification in the size and/ or the chemical composition of the component particles of the charge. It is thus possible through action on said temperature to produce for a same chemical composition of the charge, iron the composition of which varies within predetermined limits and not only, as in conventional cases, iron the composition of which is defined merely by the above-mentioned parametric factors usually resorted to.

' The following hypothesis that seems to conform with theresults obtained may explain this phenomenon.

I According to the size and chemical composition of the lumps of raw material forming the burden, the reactions between the chemical components contained therein are, more or less partial and slow, or more or less complete and rapid. When a large fraction of the more refractory substances cannot react in the solid state except through superficial spaced contacting points, whiclroccurs when the burden is constituted by large lumps and when the compounds silica, alumina and lime areessentially and separately contained in different raw materials such as alumina in bauxite, lime in limestone, the chemical reaction becomes slower and the tempera- Iture rises accordingly; this is the case for instance when thefburden contains but little silica beyond that which forms a substantial fraction of the coke ashes, because said silica contained in the coke is released only through the combustion of the coke in front of the blast furnace twyers where it reacts with the lime and the alumina contained substantially as specified above in limestone and bauxite. In contradistinction, if the burden of the blast furnace is formed of ores of silico-alumino-calcic gangue or is prepared in a manner such that the points of contact between-theditferent chemical components of the burden are numerous and multiple, the slag. forms very rapidly and the consequence is that the temperature is adjusted automatically to a lower value than in the preceding case.

The preparation of the burden that allows'obtaining multiple contact points may be effected by crushing, mixing and agglomerating the material. In practice however, coke is not admitted to such a preparation, were it only because of the permeability required for the burden.

There may appear very numerous intermediate cases between such extreme cases.

Each constituent of the burden will comprise mainly a single compound silica, alumina or lime and will be fed as large lumps, of 8 to 20 cms. for instance, when a high operating temperature is desired, alternatively as smaller pieces, i. e. less than about 8 cms., when the formation of slag is to be enhanced, thus achieving a low operating temperature of the furnace.

It is therefore possible in accordance with the invention to adjust the temperature of the blast furnace and consequently the composition of the iron, chiefly as concerns its silicon contents, by suitably choosing the nature of and the size of the lumps and pieces forming the burden.

It is also possible, in order to adjust that temperature, to increase the number of constituents of slag by introducing into the furnace appropriate additional substances capable of forming, together with the usual elements of industrial slags, complex compounds which, because of their lower melting temperatures, the eutectic law, or other properties, are more easily meltable, viz. substances such as the compounds of sodium, potassium, barium, calcium, magnesium, titanium, iron, manganese, etc. preferably as oxides, sulphides or halides. For instance the introduction (which can be easily effected through the blast nozzles) of fluorspar, in amounts 'of about 2 to 3% relatively to the amount of slag, allows of lowering by some 50 C. the operation temperature of the furnace without decreasing the flowability of the slag. Moreover these metals when added as oxides or salts, or even pure, can exert a kind of catalytic action which promotes thermodynamic equilibrium within the slag.

I have found furthermore that it may be useful to increase the time during which the slag remains in the hearth of the blast furnace so as to allow the reactions to take place and chemical equilibrium to be restored, in particular so that the lime may be combined completely with the acid radicals in the slag. This is of particular interest when burdens are used that produce a slag with a comparatively low lime content Within the above defined areas and chiefly a content that is near the lower limit of the compositions used in the execution of the invention; in such a case, it may occur that if the reactions have not actually had time to reach equilibrium, substantial fractions of the silica, of the alumina and of the lime remain combined as dicalcium silico-aluminate (gehlenite). Now the viscosity of the aluminous slag increases when the amount of such ternary compounds increases, which leads to resorting to a higher temperature of operation. Furthermore, the alumina forming part of ternary compositions such as gehlenite is generally insoluble, and cannot be recovered, whichis a further drawback when it is desired to extract the alumina from such slag.

In order to reach this result by avoiding these drawbacks, it is of advantage to increase the vertical distance between the level of the cinder notch and the level of the conventional blast nozzles so as to increase the possible volume of slag covering the iron in the hearth of the blast furnace, said slag standing between said levels, whereby it is possible for the liquid slag to remain longer at the temperature of the hearth.

In the'determination of the respective levels of the cinden notch andof the blast nozzles, shall be taken into account the fact that it is desirable to let the slag floating on the iron in the hearth during all the time separating the two successive castings of the iron which practically may be the most remote from one another in the time. As the diameter of the hearths of shaft furnaces vary considerably from one furance to another, it is not possible to give numeric values for such vertical distance. However, it may be easily determined by those skilled in the art, calling in mind that the quantitics of iron and of slag produced between said castings are well known, and consequently their volumes also, whose heights in the hearth are determined by the diameter thereof. The levels of the conventional blast nozzles and of the cinder notch should then be located in such a manner that the iron never rises to the cinder notch, and that the slag cannot reach the lower range of the blast nozzles.

Another way of achieving a longer time of presence of slag in the furnace is to avoid using, for exhaust purposes, the cinder notch, but to cast it after the iron, i. e. by substantially emptying the hearth, when the construction of the furnace, the implements at hand, and the production requirements allow such a method. In order to otbain the best possible state of equilibrium within the slag, it is generally sufiicient to keep the latter liquid for an hour inside the furnace; however, this time can be made longer. The above process, viz. the discontinuous casting of slag after the iron, through the tap hole for the iron, corresponds to periods of about 3 to 6 hrs.. when this is possible.

This way of operating brings about the following advantages which help improve the extraction yield of alumina as soluble aluminate: the slag which flows in large amounts is less cooled along its path from the tap hole to the ladle in which it collects; thus its cooling down to ambient temperature is much slower, which facilitates the allotropic transformations of the dicalcium silicate, at first from the a to the state, then from the ,3 to the 7 state, said last mentioned transformation being the main cause of the self-pulverisation of the slag by reason of the volume increase which takes place. Now this self-pulverization phenomenon is convenient because it spares costly grinding of the slag which would otherwise be required for facilitating the solution of aluminates.

For carrying out my invention on an industrial scale in a shaft furnace of any type adapted for the reducing and melting operation through which the raw iron is obtained, particularly in a blast furnace, .I operate as follows with a view to obtaining directly an iron with a high content of carbon while its sulphur content does not ever rise about 0.04% and is generally less than 0.03% and even less than 0.02%, and containing a predetermined content of silicon that is less than 1% and ranges generally between 0.2 and 0.6%.

(l) The burdens are made up by adjusting the amount of iron ore, of coke and of limestone flux, with the possible addition of aluminous material such as lateritic ore, bauxite and the like, depending on the composition of the substances, in a manner such that the slag is highly basic with high alumina content and assumes a ternary composition corresponding to a figurative point on the triangular. silica-alumina-lime ternary diagram, located inside the area MNOPQR of the ternary diagram of Fig. 1, as defined on the latter by the abovedisclosed lines, part of the lime being possibly replaced by a molecularly equivalent amount of magnesia; furthermore the usual parametric factors governing the op-. eration of the furnace are adjusted in a manner such that the temperature of the slag in the hearth, as measured at the cinder notch, assumes a predetermined value corresponding to the compositionrequired for the iron,

pper limit always lower than'1550. 'C. which is also the upper limit of theindustrial melting temperature'of the slag employed according to the invention. ;The amount of coke required for the reduction of .the ore and the melting and the carburization of the iron will be consequently calculated. for the production of iron having the above defined characteristic features.

The U. S. Patent No. 2,120,740to Faust et al. describes a process for manufacturing, in'a. shaft furnace heated with coke, a titaniferous iron containing more than 0.6% Ti (preferably from 0.8% to 1.7%) using a charge compound of bauxite, possibly titaniferous ore, limestone, coke and, if desired, iron scraps, the ratio between the weight of the slag and the weight of the iron produced being within the range of 1 to 3, said slag comprising calcium aluminate as maincomponent and the temperature in the hearth of the furnace being within the range of 1600 C. to 1800" C.

Some of theslag compositions recited in said patent (4% to 25% of SiO 35% to of CaO and 30% to of A1 0 have the same proportions of the main constituents of the slags used in the present process, and are represented on the above stated diagram by a figurative point located inside the area MNOPQR disclosed hereinabove. However, between the Fausts process, which has not the same object in view and does not produce the same iron, and the present process there is an essential diiference, viz. the temperature of the melting zone, which in said patent is always higher than 1600 C. and even higher than 1800 C. and which in the present process is always lower than 1550 C. If with the Fausts process the temperature of the melted slag is just over the industrial melting temperature t (I. M.) of said slag, with the usual margin for safety, the composition of the actual slag can only be very different from that of the slag used in the present process, because the Rankins temperature T (R.) corresponding to an industrial melting temperature of t (I. M.) 1600 C. is at least 1850 C. and even l900 C. which is well over my upper limit of 1550 C. by an amount of about 300 C.' On the contrary if in the Fausts process the used slag is strongly superheated over its industrial melting temperature, the composition of said slag may effectively have its figurative point in the area MNOPQR of the Rankins diagram but then the op eration of the furnace is quite different.

Further in the Fausts process, much more greater quantities of slag are used than in the usual operation of a shaft furnace, the weight ratio between the slag and the iron produced with said process being comprised between 1 and 3. On the contrary with the present process the weight ratio between the slag and the iron produced is within the usual range and generally has not to be higher than 1, this constitutes an important economical advantage.

Inside the above, defined area, it is of advantage to operate with a slag the figurative point of which lies inside the area m'npqr defined by the lines:

SiQ 10% SiOQ, 20% C30, A1203, CaO, 58%

The slags corresponding in composition to the points of said area have as a matter of fact a lower melting point and consequently it is possible to operate with an upper temperature limit for the slag in the hearth that is lower and is of about 1520 C.

If the composition of the iron to be obtained and the material available allow it, I may use preferably a slag the figurative point of which lies inside the composition area ABCDEF enclosed inside the preceding areas and defined hereinabove. It is then possible to operate at a temperature less than 1500 C.

(2) It is possible furthermore for a predetermined chemical composition of the slag (i. e. for a predetermined chemical composition of thercharge) to adjust as far as may be required the composition of the iron by adjusting the formation temperature of the slag :as provided by suitably, selecting the nature and the a size of the components of the charge. I introduce large lumps of a magnitude ranging between 8 and cm. for the components of the charge that are to react slowly, together with smaller pieces the size of which does not Irise above 8 cm. and which are adapted to react rapidly.

In order to obtain a temperature near the maximum limit, I introduce the components of the charge in the 'form of large lumps. In practice, the composition of ;the charge will be generally such that one is led to act chiefly on the size of the lumps of limestone and on that of the lumps of bauxite.

In order to obtain a temperature nearer the minimum limit, i. e. nearer the temperature of industrial melting of the slag corresponding to the burden that is being charged, I resort in contradistinction to small lumps in intimate admixture and it is even possible before charging to reduce through crushing or breaking, to mix or even to agglomerate all or part of the components of the charge, in order to lower as much as possible the temperature in the blast furnace, which reduces the silicon contents of the iron, the conditions of operation remaining otherwise the same.

3) It is also possible to adjust the temperature of operation by increasing the number of components of the slag by introducing into the furnace in any known manner additional substances, such as for instance compounds of sodium, potassium, barium, calcium and the like, chiefly as sulfides, fluorides and the like. It is possible to obtain in this manner, and according to the substances selected, an additional lowering of the temperature of the slag, said lowering being as considerable as ISO or even 200 C.

(4) In order to change the chemical compounds included in the slag for the purpose of lowering the viscosity of said slag, I may increase the time during which the slag remains in the hearth of the furnace, said furnace been equipped accordingly to provide that longer dura tion. The time during which the slag remains in the hearth should be preferably defined by experience so as to reach the most favorable figure, said duration depend- 12 ing on the composition, on the temperature, and on the viscosity of the slag. 7 a

(5) When, according to common practice, the blast is atmospheric air, it is useful to preheat this blast to relatively high temperatures: 700 C. and even 800 to 850 C. and even higher. However it, according to a known method, air enriched with oxygen is used, the same results 'will be achieved with a blast temperature lower than the above values, and the higher the oxygen content the lower the blast temperature that is required.

In practice, anyone skilled in the art may establish as a result of experimental runs made with the raw material available, while modifyingthe different factors disclosed hereinabove, curves showing the modifications in the composition of the iron in accordance with the variations of said factors. He may then for each application refer to these curves so as to define, for each of the above factors controlled separately or in combination, the adjustment that is most suitable for the desired result.

In order that my invention will be more fully understood the more detailed practice thereof is illustrated by the following examples:

EXAMPLES A. I have given hereinabove by way of a mere indication a few'examples of industrial runs performed in conformity with my invention.

In the following examples, the temperature of the slag inside the hearth did not rise by more than to C. above the t (I. M.) temperature referred to. In the following examples, the abbreviation T (R.) designates the temperature of incipient solidification of the slag according to Rankins diagram and t (I. M.) the industrial melting temperature of the slag as defined hereinabove.

In the analysis of the ternary composition of the slags given hereinafter, I have not taken into account the calcium combined with the sulphur as calcium sulphide nor the calcium combined as calcium titanate with the oxide of titanium existing in the slag, but only the lime available for combination with the silica and with the alumina, because, the calcium combined in said compounds is not capable of combining with the silica and the alumina of the slag, and in a first approximation all things are as if the lime corresponding to said combined calcium was inert.

Example l.- -Production of iron for steelmaking Composition of the slag Tempera- Oomposltion Analysis of the Temperature ture of the iron fraction of slag of the Analysis of the actual slag constituted by blast,

the ternary C. composition Percent O-.. 4. 72 S101.-. Si..- 0.37 A1103- Mn" 3. 04 C220 S. 0. 0010 Other elements..

Example 2.-Pr0ductzon of iron for steelmaking' Composition of the slag Tempeta- Oornposrtion Analysis of the Temperature ture of the iron traction of slag oi the Analysis of the actual slag constituted by blast,

a the ternary O.

composition G 12.70 5101.... 13.00 Si--- 31. 25 A1105--- 82. 00 Mn 0 53.50 02.0.--. 55.00 T (R.), 1,560 0-... 740 8-..- 0.009 Other elements 2.55 t (I. M.), 1,360 0-- For a duration of a few minutes only, the content of combined alumina forming a dicalcium silico-aluminate (gehlenite) was equal to 6.75%, the recoverable fraction of the alumina forming about 80 to 82%.

For a duration of 20 to 25 minutes, the first mentioned contents dropped from 6.75% to 2.9% and the yield of alumina rose to about 93%. In this case the temperature of the slag was raised to 1440 when evacuated by reason of its superheating in the hearth.

The amount of combined alumina in the shape of gehlenite has been defined by a radiographic crystalline analysis. This may in practice be estimated by testing the solubility of the alumina.

For another slag having a similar composition, the yields of alumina under comparable conditions have reached respectively 83.5% and 95%.

B. I have given hereinbelow a few examples of operative conditions that may in conformity with the invention be used with an aluminous iron ore having the following composition:

7 Percent Fe 50 Mn 0.07 S 0.1 P 0.06 SiO 2.5 A1 10.25 CaO---- 0.05 MgO 0.3 Miscellaneous The remainder The coke used is a normal metallurgical coke with l0% ash content and 1% sulphur content, the ashes containing 38% of silica, 24% of alumina, 8% of lime, 4% of magnesia and 12% of iron.

The limestone flux used contains 53% of lime, 2% of silica, 1% of alumina and 0.5% of magnesia.

The manganese ore has for its composition in the following examples:

The temperature of the blast is 750 C.

For 1000 kg. of iron containing 0.4% Si, 1.75% Mn and 4.60% C, there are obtained 740 kg. of slag containing 13.5% SiO 34% A1 0, and 52.1% CaO.

Example 12 For 1000 kg. of iron containing 0.6% Si, 1.70% Mn and 4.5% C it is necessary to charge 1,030 kg. of coke (weighed dry) 1,570 kg. of the above aluminous ore 60 kg. of the above manganese ore 16 kg. of bauxite containing 3% of SiO and 57.3% of A1203 490 kg. of ferruginous limestone containing 17.7% Fe 16 and there are obtained 720 kg. of slag containing 14% SiO 36.5% A1 0 and 49% CaO.

The temperature of the blast is 760 C.

, Example 13 In order to produce 1000 kg. of iron containing 0.4% Si, 1.79% Mn and 4.80% C, there are charged:

840 kg. of coke weighed dry 1,260 kg. of the above aluminous ore 56 kg. of the above manganese ore 525 kg. of limestone flux 357 kg. of steel scrap (turnings) There are obtained 545 kg. of slag containing 14.5% SiO 32% A1 0 and 53.5% CaO.

In the three Examples 11, 12 and 13, the temperature 1 (I. M.) of the slag, as it is being tapped, is approximately 1400 to 1450 C.

Example 14 Use of an aluminous iron ore having the following composition: a

Percent Iron 58.5 Manganese 0.4 Phosphorus 0.5

Silica 2.55 Alumina 5 Miscellaneous The remainder (This analysis is that of the aluminous ore of Mysore, India.)

With a coke (having 11% ashes) containing 3.5% alumina 4.5% silica 1% lime 0.5% magnesia 0.5% sulphur the manufacture of crude iron in the acid burdening presently used is carried out for instance with the following burden:

The above aluminous iron ore 1,610 The above co 900 The fluxes containing: Silica 87 Lime Magnesia 28 480 kg. of slag per ton of iron are obtained, this slag having the following composition:

Percent Silica 36 Alumina 24 Lime 33 Temperature of the slagin the hearth: 1500 C.

With a phosphorus content obviously equal to thatre sulting from the phosphorus in the burdens.

The process according to the inventiomappliedvuth the same ore and the same coke, leads to the following burden:

Kg. Aluminous iron ore 1,610 Coke 880 Limestone containing 236 kg. of lime.

The total amount of fiuxing agent, which is exclusively limestone, is lower than that of the previous process. It may be seen that coke consumption is decreased and that slag production is brought down to 4 -1-3 kg. per ton This slag easily flows at a temperature r (I. M.) lower than 1400 C.

The iron obtained through my process according to this example only contains 0.5% silicon and 0.01% sulphur, besides practically the same manganese and phosphorus contents as indicated above.

What I claim is:

1. In the direct manufacture in a shaft furnace of an iron having a low silicon and sulphur content, whose carbon content lies between about 4% and about 5.5%, whose silicon content lies substantially between 0.1% and 1%, and whose sulphur content is not higher than 0.04%, the process comprising melting a burden containing ferrous material selected from the group consisting of iron ores and scraps and containing coke and limestone with an amount of aluminous fiuxing agents selected from the group consisting of lateritic ores, aluminous ores and bauxites, said burden being adjusted to give a highly basic reducing and aluminous slag Whose main components are silica, alumina and lime but which may contain other components in the proportions usually present in the shaft furnace slags used in the making of crude iron, and whose ternary composition related to the combined weight of said three main components corresponds to those points on the Rankins diagram falling within the area MNOPQR of Fig. 1 of the accompanying drawing, said amount of aluminous fluxing agent being calculated to give a slag which after solidification contains pentacalcium trialuminate, working the furnace at a predetermined range of hearth temperature lower than said Rankins diagram temperature and between 1210 C. and 1500" C., and tapping the liquid iron separately from the liquid slag out of the furnace.

2. A process as in claim 1, wherein the sulphur content of the iron is 1ess than 0.02%.

3. A process as in claim 1 comprising the additional step of retaining the molten slag in the furnace during a period of at least approximately one hour and sufficient for the chemical reactions to approach equilibrium whereby the alumina contained in said slag is substantially combined in the form of binary components.

4. A process as in claim 1, wherein the slag is kept on the iron in the furnace during the whole period between successive tappings of the iron and is tapped immediately after the iron at the end of each of such periods.

5. A process as in claim 1, wherein the slag, after solidification, is essentially formed of dicalcium silicate the content of which is comprised between 23% and 63%, and of calcium aluminates, the pentacalcium trialuminate content being higher than 23%.

6. A process as in claim 1, wherein the control of the temperature of the furnace is obtained by varying the nature of the materials of the burden and/or the size thereof, large sizes of said materials ranging from approximately 8 to 20 cm. being used for obtaining higher slag formation temperatures and higher temperatures in the hearth, and small sizes below approximately 8 cm. being used for obtaining lower slag formation temperatures and lower temperatures in the hearth, whereby, for a given final composition of said slag, irons having varied compositions between the foregoing limits are obtained.

References Cited in the file of this patent UNITED STATES PATENTS Faust et al June 14, 1938 Sturbelle Sept. 24, 1946 OTHER REFERENCES UNITED STATES PATENT OFFICE CER'HFICATE m ecmmnee Patent No 2,832,682 April 29, 1958 Leonce M. Reygagne that error aptears in the above numbered patent It is hereby certified requiring correction and that the said Letters Patent should read as corrected below.

In the heading to the printed specifi insert the following:

iority, application France September 10, 1948 caticn, between lines 6 and W,

-'- Claims pr Signed and sealed this 5th day of August 1958.-

(SEAL) Attest: KARL AXLINE mime? c. wmsow Cmmieeiener of ?etente Attesting Officer

Claims (1)

1. IN THE DIRECT MANUFACTURE IN A SHAFT FURNACE OF AN IRON HAVING A LOW SILICON AND SULPHUR CONTENT, WHOSE CARBON CONTENT LIES BETWEEN ABOUT 4% AND ABOUT 5.5%, WHOSE SILICON CONTENT LIES SUBSTANTIALLY BETWEEN 0.1% AND 1%, AND WHOSE SULPHUR CONTENT IS NOT HIGHER THAN 0.04% THE PROCESS COMPRISING MELTING A BURDEN CONTAINING FERROUS MATERIAL SELECTED FROM THE GROUP CONSISTING OF IRON ORES AND SCRAPS AND CONTAINING COKE AND LIMESTONE WITH AN AMOUNT OF ALUMINOUS FLUXING AGENTS SELECTED FROM THE GROUP, CONSISTING OF LATERITIC ORES, ALUMINOUS ORES AND BAUXITES, SAID BURDEN BEING ADJUSTED TO GIVE A HIGHLY BASIC REDUCING AND ALUMINOUS SLAG WHOSE MAIN COMPONENTS ARE SILICA, ALUMINA AND LIME BUT WHICH MAY CONTAIN OTHER COMPONENTS IN THE PROPORTIONS USUALLY PRESENT IN THE SHAFT FURNACE SLAGS USED IN THE MAKING OF CRUDE IRON, AND WHOSE TERNARY COMPOSITION RELATED TO THE COMBINED WEIGHT OF SAID THREE MAIN COMPONENTS CORRESPONDS TO THOSE POINTS ON THE RANKIN''S DIAGRAM FALLING WITHIN THE AREA MNOPQR OF FIG. 1 OF THE ACCOMPANYING DRAWING, SAID AMOUNT OF ALUMINOUS FLUXING AGENT BEING CALCULATED TO GIVE A SLAG WHICH AFTER SOLIDIFICATION CONTAINS PENTACALCIUM TRIALUMINATE, WORKING THE FURNACE AT A PREDETERMINED RANGE OF HEARTH TEMPERATURE LOWER THAN SAID RANKIN''S DIAGRAM TEMPERATURE AND BETWEEN 1210*C. AND 1500*C., AND TAPPING THE LIQUID IRON SEPARATELY FROM THE LIQUID SLAG OUT OF THE FURNACE.

Images