Method relating to manufacturing of steel

Abstract

The invention concerns a method for the manufacturing of steel in an electric arc furnace, comprising melting of charged steel raw material, substantially iron carrier, characterised in that at least 5 weight-%, preferably at least 10 weight-%, of charged iron carrier consist of granulated pig iron, here denominated GPI.

Description

TECHNICAL FIELD

[0001] The invention concerns a method relating to manufacturing of steel in an electric arc furnace, comprising melting charged steel raw materials for steel manufacturing.

BACKGROUND OF THE INVENTION

[0002] The dominating steel raw material for manufacturing of steel in electric arc furnaces is scrap. In year 1994 374.2 millions tons of steel were produced in electric arc furnaces, of which 93.5% were produced by remelting scrap. Remaining steel raw materials for steel making mainly consisted of pig iron and directly reduced iron, herein denominated DRI.

[0003] DRI is manufactured through a number of different processes, among which Midrex-process is the dominating technique. Other employed techniques of those, which are known by the denominations HYL, HYLIII, FIOR and FASTMET (trade name). Iron carbide Fe3C, is another product, which to a limited degree is available as a substitute to scrap.

[0004] Table 1 shows the typical range of composition of DRI along with the data of Fe3C.

[0005] The most pronounced advantage of using DRI/Fe3C materials is the low content of residuals (Cu, Sn, etc.), which generally are considered as harmful, which opens for the possibility to dilute scrap of poor quality. The DRI/Fe3C also is relatively high in carbon that results in CO (g) formation when oxygen is injected into the steel. The CO (g) will reduce the steel nitrogen content and enhance slag foaming. In addition, the consistence of the DRI/Fe3C composition with time, enables the EAF operator to have a smooth process with small alterations between different heats. 1 TABLE 1 Typical chemical composition of DRI and iron carbide. % DRI Iron Carbide Fetot 87-94 89.8 Femet 76-89  1.0 Fe3C — 90.0 C 0.2-2.4  6.0 S 0.01-0.03 — P 0.007-0.05  — SiO2 + Al2O3 2.6-6.7  2.4 CaO − MgO 0.2-3.0 —

[0006] The DRI/Fe3C, however, also causes some negative effects to the Electric Arc Furnace (EAF) process compared to scrap. Relatively high gangue and iron oxide levels result in a higher energy demand. An estimation shows that each additional percent of oxygen that replaces 1% of iron will cost 49 kWh/ton, which in turn results in increased electrode consumption, tap-to-tap time and requires an additional amount of carbon of 6.8 kg/ton. Altogether, producing liquid steel from DRI requires 240 kWh/ton more than producing the same steel from scrap.

[0007] Consequently, it is evident that the use of DRI/Fe3C in the EAF is most common where the production of DRI/Fe3C is cheap, i.e. where natural gas is commonly available, usually in combination with lack of high quality scrap and/or production of residual sensitive steel grades, predominantly in several developing countries, where the use of DRI may represent 10-100% of the charged material in some EAF:s.

[0008] A more readily available scrap substitute than DRI/Fe3C is pig iron. In fact, pig iron is already today charged in many EAF:s, wherein the pig iron consists of conventional shapes produced in pig iron casting machine, sand lined pit casing or the like. These pig iron shapes, however, are not designed to fit the requirements of an EAF steel raw material very well, and particularly it does not promote the control of the melting and decarburisation and reduction processes which are carried out in the EAF.

BRIEF DESCRIPTION OF DRAWINGS

[0009] In the following description of the invention, reference will be made to the accompanying drawings, in which

[0010] FIG. 1 in the form of a diagram illustrates how the content of residual metals emanating from scrap can be reduced at the manufacturing of steel according to the invention, and

[0011] FIG. 2 and FIG. 3 in the form of diagrams illustrate the effects of pre-heating and post-combustion, respectively.

DISCLOSURE OF THE INVENTION

[0012] It is the purpose of the invention to provide an improved method relating to manufacturing of steel in an EAF (Electric Arc Furnace) comprising melting charged steel raw materials for steel manufacturing and preferably also decarburisation by injection of oxygen gas in the molten metal. The invention is characterised in that x-100 weight-% of charged steel raw materials for steel manufacturing consists of granulated pig iron, herein denominated GPI. Preferably said GPI satisfies the following requirements, namely:

[0013] a) that it has a chemical composition containing 0.2-3% Si, 2-5% C, 0.1-6% Mn, the remainder essentially only iron and impurities which can normally exist in pig iron produced in the blast furnace process or other shaft furnace process, e.g. in Capola furnace,

[0014] b) that it has a melting point <1350° C., and

[0015] c) that it consists of essentially homogenous particles with substantially round or oval shape obtainable through granulation of a melt with the above mentioned composition, comprising disintegration of a stream of said melt to drops, which are cooled in a water bath to form a granulate.

[0016] For the production of the granulate, a specific method can be used, the general principles of which are described in the U.S. Pat. No. 3,888,956. By this known method, raw iron granulate can be produced, in which at least 90 weight-% of the granulates consist of particles with shapes varying from substantially round or oval disks to drops and spheres with sizes varying from 1 mm up to 25 mm measured in the largest dimension of the granules. The GPI can be used in this form, but preferably the fine fraction is removed by screening (this finer fraction can be used as a doping agent in a foaming slag in the EAF, as will be explained more in detail in the following), so that the GPI which in accordance with what is mentioned above is charged to form a melt an/or which is added to a form or remaining melt consists of a granulate which to at least 80 weight-% consists of particles having a particle size between 2 mm and 25 mm, measured in the largest dimension of the granules.

[0017] The low area/volume ratio of the round or oval particles of the GPI reduces oxidation during storage and handling, something which has turned out to be a problem with DRI with its porous structure. The area/volume ratio of GPI, however, is higher than of normal pig iron and large sized scrap material, and considerably more well defined, which provides a better and more reproducible heating and melting features. The round or oval shape of the GPI also results in a relatively high bulk density, approximately 4.5 kg/l, with excellent free-flow characteristics. Most commercial scrap grades such as bundles, shred metal and turnings, have a bulk density of 0.7-1.2 kg/l, table 1. The GPI's shape also enables easy penetration through the slag layer when the iron is injected into the EAF. Finally the GPI, when screened as above indicated, has a low fraction of fines and is relatively hard, which in combination gives small material losses during handling. 2 TABLE 2 Bulk density and residual levels of GPI, DRI and some scrap grades Bulked density (kg/l) % (Cu + Ni + Mo + Sn) Granulated Pig Iron 4.5 0.05 (GPI) Direct Reduced Iron  16-2.7 0.05* (DRI) No 1 Bundles. 1.1 0.16 No 2 Bundles 0.85-1.0  0.72 No 1 HMS 0.7-0.8 0.50 No 2 EMS 0.65-0.8  1.01 Shredded 1.0-1.1 0.56 *Includes the elements Pb, Sn, Mo, Zn, W, As, Sb, Co, Ni and Zr.

[0018] According to a preferred embodiment of the invention, there is charged to the EAF as a steel raw material for making steel, besides GPI also scrap containing impurities in the form of one or more of the metals which belong to the group of residual metals which consist of e.g. copper, nickel, molybdenum, and tin. Therein there is achieved an advantageous dilution of said residual metals in the finished steel melt because the GPI contains significant lower levels of residuals (Cu, Sn, Ni, etc.) than scrap, table 1. The dilution effect of GPI addition to the EAF on the residual content is illustrated in FIG. 1.

[0019] The low residual levels of GPI opens for the possibility of the EAF operator to use poorer scrap quality, FIG. 1. Typically, when using a mixture of GPI and scrap according to the invention, the addition of GPI amounts to at least 10%, preferably more than 25%, or even more than 40% of the added steel raw materials, the remaining steel raw material being substantially scrap. However, addition of GPI as the sole, 100% steel raw material can be contemplated, particularly when producing steel intended for flat products for which virgin steel raw material is particularly advantageous.

[0020] GPI has a higher content of carbon compared to scrap and DRI. When the carbon is decarburised by oxygen injection, the CO (g) purging reduces the nitrogen level of the steel and chemical heat is generated. Apart from rinsing the steel, the CO formation during oxygen injection may be used in order to form a foamy slag. If a carbon injection of 12 kg/ton steel is used during normal operation for this matter, approximately 30-40% of the charged material can be substituted by GPI only in order to balance the carbon injection. An additional benefit of adding the carbon as GPI instead of injected carbon, is the possibility of achieving an early boil, i.e. GPI opens for an early slag foaming, which increases the heat efficiency and eliminates any power reduction due to thermal overload.

[0021] The GPI chemical composition also differs with respect to some additional properties compared to scrap, DRI and Fe3C. Thus, GPI has a very low oxide content. DRI/Fe3C, on the other hand, contains a rather large amount of gangue and unreduced iron oxides, which require additional energy to be added. Further, GPI is relatively high in silicon. This silicon is oxidised during melting and oxygen injection and requires an extra lime addition in order to control the slag composition. This lime requires extra energy input in order to heat and melt the slag former. In the case of GPI, however, the extra energy need is by far compensated for by the chemical heat evolved during silicon oxidation and may even allow the addition of DRI together with GPI without need for additional electrical power supply compared to a case of 100% scrap charge. According to one aspect of the invention therefore there is added, besides GPI to the EAF, also directly reduced iron, DRI, which contains in weight-% 75-90% metallic iron, 0.2-3% C, 2-7% gangue material, mainly SiO2+Al2O3, the balance being substantially iron oxide, FeO, wherein GPI is added at least in an amount such that its content of silicon and carbon in combination with the carbon in added DRI will reduce the iron oxide in said DRI to form elementary iron, at the same time as the oxidation of silicon and carbon in said GPI and DRI generates heat to a sufficient degree for compensating the cooling effect that is caused by the gangue material and the iron oxide in added DRI and preferably also compensates for the cooling effect because of added lime or other basic slag former (Mg- and/or Ca-carrier) for controlling the slag composition.

[0022] The GPI can be added through basket charging as well as by continuous feeding, e.g. via a vertical scrap chute or by injection. When basket charged, the GPI should be added in the first basket that is charged in the EAF, wherein a melt is quickly formed because of the low melting temperature of the GPI. When adding by injection, the GPI addition may eliminate at least one basket of scrap. This will decrease the furnace idle time as well as heat losses. In addition, continuous feeding of material into the EAF results in a much smoother operation of the furnace, compared to batch-wise addition of scrap. At continuous feeding, foaming slag practice in combination with maximum power input can be applied. The high bulk density of GPI also is an advantage when basket charging is used only. In conclusion, the possibility of continuously adding steel raw material into the EAF is from a practical point of view a strong argument to use GPI.

[0023] The rather high carbon content of GPI results in a low melting point, that is an early melting is achieved in the furnace. Once a liquid steel pool is present in the furnace, the GPI injection can start through the furnace roof. If then a foamy slag is formed on top of the steel and constant (maximum) electrical power is applied, the evolved heat from the electric arcs may be balanced by the injection rate of GPI. This will open for the possibility of keeping the steel temperature constant and minimises the thermal gradients in the furnace volume, one of the drawbacks in a “normal EAF”.

[0024] Temperature control of the steel during melting also increases the possibilities of performing refining operations at an early stage in the furnace and it opens for the possibility of running the EAF semi-continuously, that is with a rather large hot heel, continuous feeding of steel raw material and batch-wise bottom tapping.

[0025] If an even more energy efficient production is desired, the CO (g) formed during slag foaming can be subject to post-combustion above the steel bath. In addition, the GPI can be preheated by the furnace exhausts to high temperatures without any risk of environmentally hazardous emissions, which even more increases the heat efficiency of the furnace. These issues are further discussed below.

[0026] In order to illustrate the thermal benefits of using GPI, the energy needed for melting and heating is calculated on the material types shown in table 3. The calculations are based on the assumption that all C and Si are subject to oxidation, the CO formed is not post-combusted and all formation of SiO2 during melting is assumed to be neutralised by the addition of CaO or other neutralizing agent. The energy required for melting of CaO or corresponding and the energy evolved when CaO, SiO2 and other oxides are mixing, is assumed to be equal.

[0027] Table 3 also gives the theoretical energy requirement in order to melt and superheat the materials to a temperature of 1600° C. Given figures are per ton produced pure Fe. As can be seen from the table, GPI requires the lowest amount of electrical energy due to the latent chemical heat available in the material. It can also be understood from table 3 that the rather large difference between GPI # 1 and #2 is due to the difference in % Si; 0.5 and 1.2 respectively. 3 TABLE 3 Material compositions and energy requirements for various materials when aiming at different slag basicities. DRI GPI Pure Fe (Midrex) Fe3C #1 #2 Composition % C 0 1.5 0 4.2 4.6 % Si 0 0 0 0.5 1.2 % FeO 0 6 6.6 0 0 % SiO2 0 1.5 1.2 0 0 % Al2O3 0 0.8 1.2 0 0 % CaO 0 1.2 0 0 0 % MgO 0 0.3 0 0 0 % Fe3C 0 0 90 0 0 % Fe 100 88.7 1 95.3 94.2 [KWh/ton B = 1.2 378 438 335 266 217 pure Fe] B = 1.5 378 439 337 268 220 B = 1.8 378 440 338 269 223

[0028] The possibility of utilising scrap preheating is illustrated in FIG. 2, which presents a calculation of the energy need at different degrees of preheating for the table 3 materials. It should be noticed that the preheating of scrap is limited to 300° C. due to environmentally reasons. Preheating of DRI might also be restricted due to the pyrophoric behavior.

[0029] If post combustion is carried out in the EAF, the formed CO (g) during decarburisation can be oxidised and the latent chemical heat of GPI will be even more efficiently used. FIG. 3 shows the theoretical energy requirements versus the amount of post combusted CO (g) that forms CO2 (g) (100% yield of produced heat). Added material is preheated to 200° C.

[0030] The invention is particularly suited to be employed for the manufacturing of steel in an EAF (Electric Arc Furnace) comprising the formation of a foaming top slag with a temperature of 1400-1800° C. in the furnace on top of the surface of the molten metal and supply of oxygen gas to the molten metal in order to oxidise at least part of the existing silicon in the melt for heat generation and to oxidise at least part of the carbon in the melt for heat generation and to generate gas in the form of CO and/or CO2 which contributes to the slag foaming, by which the supply of oxygen to the melt also brings about oxidation of metal elements other than silicon in the melt, in this text generally referred to as valuable metal elements, which go into the slag and are reduced there by the addition of reduction agents to the top slag so that these elements to a considerable degree are recovered to the melt. According to this aspect of the invention, there is, during at least one phase of the production process, a doping agent in the form of a particle-formed, granulated product is added to the top slag with the aim of creating improved conditions for the reduction of the oxidised, valuable metal elements in the top slag, participating in the reduction process itself, contributing to and/or maintaining the slag foaming as well as adding metal to the melt, said doping agents fulfilling the following requirements, namely:

[0031] a) that it has a chemical composition containing 0-5% Si, 2-7% C, 0-3% Mn, the remainder essentially only iron and impurities which can normally exist in pig iron produced in the blast furnace process or other shaft furnace process,

[0032] b) that it has melting point <1350° C., and

[0033] c) that it consists of essentially homogeneous particles with substantially round or oval shape obtainable by granulation of a melt with above-mentioned composition, comprising disintegration of a stream of said melt to drops, which are cooled in a water bath to form a granulate.

[0034] Preferably the said doping agent is of the same general type that is used as a steel raw material and which is melted to form a bath of molten metal as has been described in the foregoing. Preferably, however, the GPI, which is added as a doping agent to the slag, has a smaller particle size than the GPI that is added as a steel raw material to the furnace according to the foregoing. More particularly, it is advantageous to use, as doping agent, GPI which to at least 80 weight-% consists of particles having a particle size varying between 0.5 mm and 5.5 mm measured in the largest dimension of the particles. The doping agent thus may consist of a fine fraction of granulated pig iron, the main part of the granulate having a considerably larger particle size, 2-25 mm, being basket charged to the furnace or injected into the melt that is successively formed. A granulate having said smaller particle sizes has a capacity to penetrate the slag to a desired degree and to keep themselves suspended in the slag for sufficiently long not only to melt, which the particles do quite quickly, but also in order that the content of carbon and silicon of the granulate shall get sufficient time to react with the oxides of the valuable metal elements in the slag, and successively agglomerate to form larger agglomerate of molten metal, which sink down through the slag to combine with the melt. The rounder the particles are, the better they are in terms of their ability to penetrate the slag. In contrast, irregularly shaped ships, flakes, oxide scales, etc. have very poor penetration ability, which is also true for powder, and cause large losses to the flue system when injected.

[0035] The addition of doping agent can be made through a lance with a gas carrier, where the lance can be placed through the slag door, furnace wall or furnace roof, or by mechanical feeding from a position above the slag, in the furnace wall or furnace roof. The added doping particles melt quickly in the hot slag and form small drops with large boundary layer area between liquid metal phase and slag, which kinetically favours reduction of metallic oxides. The doping agent contains active contents of dissolved carbon and silicon, which participate as melted drops in the reduction reactions. Dissolved carbon forms CO/CO2-gas, which in turn generates and/or maintains the foaming slag and helps to keep the small metal drops suspended in the slag. The achieved reduction and foaming implies a number of advantages in the process, which in certain cases can be vital for obtaining an economically acceptable furnace operation. Thus, the carbon dissolved in the doping agent has several functions: it contributes to and/or maintains formation of the foaming slag, it contributes to keeping the small molten metal drops suspended in the slag, which maintains the foaming, and it participates in the reduction processes.

[0036] Also the silicon, which is dissolved in the doping agent, has several functions. Silicon contributes to the reduction of oxidized valuable metal elements, which most probably decreases the boundary layer tension between slag and doping agent, which further accelerates the reduction reaction. Furthermore, heat is formed through the oxidation of dissolved carbon and silicon. Oxidation of dissolved silicon contributes as well to the formation of slag in the furnace. Finally, the doping agent contributes to a significant addition of iron to the melt when most of the reduction agents—C and Si—in the doping agent have reacted with the slag and a number of smaller drops have agglomerated to larger drops which then sink down through the slag layer into the metal bath.

[0037] In order for the doping agent to be useful as a commercial product of high value, with which the electric arc furnace top slag can be doped to produce the desired result from heat to heat, it is desirable that the contents of carbon and silicon in the doping agent be kept within relatively narrow limits within the stated outer limits. Thus the contents of carbon and silicon should not vary more than +/−0.5%, preferably not more than +/−0.3% from the assigned target value within said outer limits. Thus the carbon content in the doping agent should go up to (Cx+/−0.5)%, where Cx is a number between 3 and 4.5. Preferably the carbon content should be (Cx+/−0.3)%. In a corresponding way the silicon content should be (Six+/−0.5)%, preferably (Six+/−0.3)%, where Six is a number between 1 and 2.5. The desired contents of carbon and silicon can be obtained through alloying the raw iron with carbon and silicon after possible desulphurisation or other treatment of the raw iron.

[0038] The amount of added doping agent can be varied within wide limits depending on the composition of the melt, the composition of the doping agent and other factors. Normally the amount of doping agent added to the slag according to the invention can go up to between 5 and 60 kg of doping agent per ton produced steel, which is added to the slag by injection into the slag or in another manner to maintain slag foaming and reduction. Simultaneously, oxygen is added in a balanced amount to the steel to oxidize mainly Si and C in the steel to obtain heat and gas for the slag foaming. Also other metal elements in the steel, e.g. Fe and Cr, are oxidized to a certain degree and then reduced again when they reach the slag. Other reduction agents may also be added, besides the doping agent according to the invention, e.g. C or Si, to the slag in order to ensure the reduction together with the doping agent according to the invention. Preferably, the reduction agent, however, completely consists of the doping agent of the invention, which is advantageous from several reasons. On one hand the doping agent of the invention has evident economical merits; on the other hand the process is easier to control if the number of different additions is restricted.

[0039] The GPI which is employed according to the invention, in the first place for forming a bath of molten metal, which GPI is basket charged or is injected in the melt, as well as the GPI which possibly is injected as a doping agent in connection with foamed slag practice, can be manufactured according to a number of different methods, comprising granulation of a pig iron melt having the above mentioned composition, comprising disintegration of a stream of molten metal to drops, which are cooled in a water bath to form a granulate. A useful technique, which is known under the trade name GRANSHOT is as mentioned described in the U.S. Pat. No. 3,888,956, which also describes how the size and shape of the granulate that is being manufactured can be controlled through variation of the height of fall of the stream of molten metal before the stream is disintegrated to drops and/or of the height of fall of the drops before they hit the water surface in the cooling bath. In addition and/or as an alternative the achieved granulate may be a sieved for the provision of the desired size fraction or desired size fractions, respectively, according to above.

[0040] Concerning GPI as a charging material, i.e. as a steel raw material—in which above mentioned doping agent is not included—in EAFs, the following applies, which can be utilised according to different aspects of the invention:

[0041] 1. GPI replaces completely or partly other steel raw materials, such as for example scrap, conventional pig iron, DRI and/or Fe3C, and can according to an aspect of the invention be charged through the use of scrap baskets and/or equipment for continuous charging.

[0042] 2. GPI has very good preheating features, especially for preheating by the flues from the furnace because of the shape and chemical stability of the GPI, which is taken advantage of according to another aspect of the invention, which characterised in that the GPI that is charged in the furnace is preheated by the flue (exhaust) gases of the furnace before charging. This possibility does not exist with scrap, at least not to the same degree, because of the shape of scrap and also because of the risk of formation of dioxine, or with DRI because of the pyrophorous character of that material

[0043] 3. GPI has good “free flow” features, which facilitates continuous charging, which can be employed according to an other aspect of the invention. Herethrough a plurality of essential advantages can be gained, such as

[0044] Less heat losses because of turned-aside furnace roof in connection with basket charging.

[0045] Shorter power-off time.

[0046] Smoother procedure of process (=higher energy yield) when the arcs continuously and in, a stabile manner hit against a steel bath.

[0047] Possibility of a continuous foamed slag process, i.e. higher yield of supplied energy.

[0048] Possibility to control the temperature of the steel during the melting phase, which means that refining operations may be initiated during an early phase.

[0049] Higher productivity because of shorter tap-to-tap times, i.e. shorter times of treatment reduce all time-dependent losses.

[0050] Possibility to run the furnace continuously or semi-continuously, which further increases the productivity of the furnace.

[0051] 4. GPI has a high bulk density, which is an advantage during all phases of the handling of the material, from transportation to charging.

[0052] 5. GPI has low melting point (<1350° C.) which gives an early metal melt in the furnace, a possibility which is also utilised according to a number of further aspects of the invention, which are described in the foregoing, in the appending patent claims and/or below.

[0053] 6. The comparatively high contents of C and Si make oxygen gas injection in the molten metal possible, and hence chemical heat when CO/CO2 and SiO2 are formed, without any greater risk of oxidation of other alloy elements. The latter is often the case according traditional technique when charged Si does not exist in the steel until after a certain period of time, when oxidation of the alloying elements already have taken place. Oxygen gas injection also means that a liquid, preferably foaming slag, can be created at an early stage of the process.

[0054] 7. If oxygen gas is added to the melt, which is the case according to a preferred embodiment of the invention, the oxidation of C and Si in GPI means a very low consumption of energy for melting GPI, wherein the heat which is generated through the oxidation of C and Si also can be used as a contribution to melting any possibly added scrap, and/or DRI and/or for compensating for heat losses because of reduction of iron oxide in any possibly added DRI, addition of lime or other basic slag former for controlling the basicity of the slag, etc.

[0055] 8. Some advantageous chemical features of GPI are, besides the high contents of C and Si as above mentioned, also the following:

[0056] Law contents of residuals (Cu, Sn, Zn, Ni, etc.) which according to an aspect of the invention can be utilised e.g. for diluting the content of such residual elements in the molten metal, when also scrap is used as a steel raw material in the charge.

[0057] Very small or no fraction of oxidic material.

[0058] Homogenous composition; no differences of practical importance within the same batch or from batch to batch.

[0059] Particularly advantageous is the invention when GPI is used as a charging material, preferably in the form of a coarse fraction of a pig iron granulate according to above (80 weight-% of the GPI having particle sizes between 2 and 25 mm), as well as a doping agent in foaming slag, preferably in the form of a fine fraction of pig iron granulate (80 weight-% having sized between 0.5 and 5.5 mm). When GPI is charged and is melted in the furnace, a C-and Si-buffer is ensured in the melt. If the charging of GPI to the melt further is performed continuously, the oxidation of alloying elements is further diminished. This creates conditions for valuable metals, in the first place iron and possibly other existing, oxidised metals to be recovered by reduction through addition of only GPI in its function as a doping agent to the slag for securing a complete slag reduction.

Claims

1. Method relating to manufacturing of steel in an electric arc furnace comprising melting charged steel raw material, substantially iron carrier, characterised in that at least 5 weight-%, preferably at least 10 weight-%, of charged iron carrier consists of granulated pig iron, herein denominated GPI.

2. Method according to claim 1, charecterised in that said GPI satisfies the following conditions, namely:

a) that it has a chemical composition containing 0.2-3% Si, 2-5% C, 0.1-6% Mn, the remainder essentially only iron and impurities which can normally exist in pig iron produced in the blast furnace process or other shaft furnace process, e.g. in Capola furnace,

b) that it has a melting point<1350° C., and

c) that it consists of essentially homogenous particles with substantially round or oval shape obtainable by granulation of a melt with the above mentioned composition, comprising disintegration of a stream of said melt to drops, which are cooled in a water bath to form a granulate.

3. Method according to claim 2, characterised in that it comprises decarburisation through oxygen gas injection into molten metal formed in the furnace.

4. Method according to claim 3, characterised in that the silicon in said GPI is oxidised at the decarburisation to silicon dioxide, SiO2, which essentially is collected in a top slag in the furnace, for the control of the slag composition, wherein there is added any basic slag former, substantially containing Ca- and/or Mg-carriers in such an amount that the slag composition will satisfy the requirement

1 2,   ⁢ 8 ≤ CaO + MgO SiO 2 ≤ 3,   ⁢ 6, ⁢  

preferably the requirement

2 3,   ⁢ 1 ≤ CaO + MgO SiO 2 ≤ 3,   ⁢ 3

5. Method according to any of claims 1-3, characterised in that the steel is produced batch-wise in the electric arc furnace and that said GPI is added to the electric arc furnace at an initial stage of the charging procedure in order quickly to form a pool of molten metal in the furnace.

6. Method according to claim 5, characterised in that said steel raw material at least partly is basket charged, at least GPI being added with the first basket in the charging procedure.

7. Method according to claim 5 or 6, characterised in that said GPI is injected in the pool of molten metal which initially is formed or added and/or in the pool of molten metal that successively is formed in the furnace.

8. Method according to any of claims 1-4, characterised in that the furnace is operated semi-continuously, i.e. with batch-wise bottom-tapping of a portion, preferably 40-60% of the steel melt, and that GPI is charged continuously or semi-continuously to remaining pool of molten metal and/or to the successively growing pool of molten metal.

9. Method according to any of claims 1-8, characterised in that said GPI to at least 80 weight-% consists of particles having a particle size between 2 mm and 25 mm measured in the largest dimension of the particles.

10. Method according to any of claims 1-9, characterised in that said GPI has a bulk density of 3.5-5.5, preferably 4-5 kg/l.

11. Method according to any of claims 1-10, characterised in that said GPI is preheated by the flue gases from the furnace before charging, preferably that said GPI is preheated continuously by the flue gases before continuous or semi-continuous charging.

12. Method according to an of claims 1-11, characterised in that as a steel raw material there is added to the furnace, besides said GPI, also scrap which contains impurities in form of one or more of the residual metals belonging to the group of metals consisting of Cu, Ni, Mo, Zn and Sn, wherein the addition of said GPI dilutes the content of said residual metals in the steel melt being formed.

13. Method according to any of claims 1-11, characterised in that as a steel raw material there is added to the electric arc furnace, besides said GPI, also directly reduced iron, here denominated DRI, which contains in weight-% 75-90% metallic iron, 0.2-3% C, 2-7% gangue material, substantially SiO2+Al2O3, the balance being substantially iron oxide, FeO (iron bound as oxides), wherein GPI is added at least in such extent that its content of silicon and carbon in combination with carbon in added DRI will reduce the iron oxide of said DRI to metallic iron, at the same time as the oxidation of Si and C in said GPI generates heat at least in a sufficient amount to compensate for the cooling action caused by the gangue material and the iron oxide in added DRI.

14. Method according to claims 12 and 13, characterised in that as a steel raw material there is added to the furnace, besides GPI, also scrap containing impurities in the form of one or more of the residual metals belonging to the group of metals consisting of Cu, Ni, Mo, Zn and Sn, wherein the addition of said GPI will dilute the content of said residual metals in the steel melt that is being formed, wherein also directly reduced iron being added, herein denominated DRI, containing in weight-% 75-90% metallic iron, 0.2-3% C, 2-7% gangue material, substantially SiO2+Al2O3, the balance being substantially iron oxide, FeO (iron bound as oxides), wherein GPI is added at least in such extent that its content of silicon and carbon in combination with carbon in added DRI will reduce the iron oxide of said DRI to metallic iron, at the same time as the oxidation of Si and C in said GPI generates heat at least in a sufficient amount to compensate for the cooling action caused by the gangue material and the iron oxide in added DRI.

15. Method according to any of the previous claims, characterised in that 10-20%, preferably 30-50% of the steel raw material consists of said GPI.

16. Method according to claim 1, characterised in that steel raw material consists of said GPI to 100%.

17. Method according to claim 1, comprising the formation of a foaming slag with a temperature of 1500-1750° C. in the furnace on top of the surface of the bath of molten metal, and the supply of oxygen in the form of oxygen gas to the melt to oxidise at least part of carbon existing in the melt for heat generation and to generate gas in the form of Co and/or Co2 as a contribution to the slag foaming, wherein the supply of oxygen to the melt also brings about oxidation of other metal elements than silicon in the melt, herein referred to as valuable metal elements, which enter the top slag from where they at least to an essential degree are recovered to the melt through addition of reduction agents to the top slag, characterised in that during at least one phase of the one phase of the production process, a doping agent in the form of a particle-formed, granulated product is added to the top slag with the aim of creating improved conditions for the reduction of the oxidised, valuable metal elements in the top slag, participating in the reduction process itself, contributing to and/or maintaining the slag foaming as well as adding metal to the melt, said doping agents fulfilling the following requirements, namely:

a) that it has a chemical composition containing 0-5% Si, 2-7% C, 0-3% Mn, the remainder essentially only iron and impurities which can normally exist in pig iron produced in the blast furnace process or other shaft furnace process,

b) that it has melting point<1350° C., and

c) that it consists of essentially homogeneous particles with substantially round or oval shape obtainable by granulation of a melt with above-mentioned composition, comprising disintegration of a stream of said melt to drops, which are cooled in a water bath to form a granulate.

18. Method according to claim 17, characterised in that the particles which are added to the slag consist of particles which to at least 80 weight-% consist of particles having a particle size varying between 0.5 and 5.5 mm measured in the largest dimension of the particles.