Computer simulation of equilibrium relations in managanese ferroalloy production

The complex heterogeneous equilibria associated with the production of manganese ferroalloys have been simulated using recently developed thermochemical databases. Over 600 measured equilibrium data have been used to verify and calibrate the model calculations. Good agreement is obtained for many important technical relations such as carbon solubility, equilibrium diagrams and phase relations, and element distribution as a function of temperature and composition. Experimental slag/metal and slag/metal/gas equilibria in MnO-SiO₂ binary, MnO-SiO₂-CaO and MnO-SiO₂-Al₂O₃ ternary, MnO-SiO₂-CaO-Al₂O₃ quaternary, and MnO-SiO₂-CaO-Al₂O₃-MgO quinary systems can be reproduced within experimental uncertainties. The influence of temperature, CO partial pressure, and slag chemistry on the Mn- and Si-distribution equilibria has been quantitatively evaluated. This leads to a sound basis for optimizing the manganese ferroalloy production processes.     

The behaviour of chromium in reduced slag-metal systems

The equilibrium between slag and metal in the CaO-SiO₂-Al₂O₃-CrO_x-FeO/Fe-Cr-Si system was investigated and the results expressed in terms of the apparent equilibrium constants for the Cr/FeO and Cr/SiO₂ equilibria. The effect of slag composition and temperature on the apparent equilibrium constants is described. The predominant form of chromium in the slags was CrO and the thermodynamic properties of CrO in the slag are discussed. The equilibrium results are applied to the stainless steelmaking system in order to show the effect of slag composition, silicon content of metal and temperature on the equilibrium chromium content of the slag.     

Experimental Investigation of Gas/Slag/Matte/Tridymite Equilibria in the Cu-Fe-O-S-Si System in Controlled Atmospheres: Development of Technique

The majority of primary pyrometallurgical copper making processes involve the formation of two immiscible liquid phases, i.e., matte product and the slag phase. There are significant gaps and discrepancies in the phase equilibria data of the slag and the matte systems due to issues and difficulties in performing the experiments and phase analysis. The present study aims to develop an improved experimental methodology for accurate characterisation of gas/slag/matte/tridymite equilibria in the Cu-Fe-O-S-Si system under controlled atmospheres. The experiments involve high-temperature equilibration of synthetic mixtures on silica substrates in CO/CO₂/SO₂/Ar atmospheres, rapid quenching of samples into water, and direct composition measurement of the equilibrium phases using Electron Probe X-ray Microanalysis (EPMA). A four-point-test procedure was applied to ensure the achievement of equilibrium, which included the following: (i) investigation of equilibration as a function of time, (ii) assessment of phase homogeneity, (iii) confirmation of equilibrium by approaching from different starting conditions, and (iv) systematic analysis of the reactions specific to the system. An iterative improved experimental methodology was developed using this four-point-test approach to characterize the complex multi-component, multi-phase equilibria with high accuracy and precision. The present study is a part of a broader overall research program on the characterisation of the multi-component (Cu-Fe-O-S-Si-Al-Ca-Mg), multi-phase (gas/slag/matte/metal/solids) systems with minor elements (Pb, Zn, As, Bi, Sn, Sb, Ag, and Au).     

Experimental Investigation of Gas/Slag/Matte/Tridymite Equilibria in the Cu-Fe-O-S-Si System in Controlled Gas Atmosphere: Experimental Results at 1523 K (1250 °C) and <Emphasis Type="Italic">P</Emphasis>(SO<Subscript>2</Subscript>) = 0.25 atm

To assist in the optimization of copper smelting and converting processes, accurate new measurements of the phase equilibria of the Cu-Fe-O-S-Si system have been undertaken. The experimental investigation was focused on the characterization of gas/slag/matte/tridymite equilibria in the Cu-Fe-O-S-Si system at 1523 K (1250 °C), P(SO₂) = 0.25 atm, and a range of P(O₂)s. The experimental methodology, developed in PYROSEARCH, includes high-temperature equilibration of samples on substrate made from the silica primary phase in controlled gas atmospheres (CO/CO₂/SO₂/Ar) followed by rapid quenching of the equilibrium condensed phases and direct measurement of the phase compositions with electron-probe X-ray microanalysis (EPMA). The data provided in the present study at 1523 K (1250 °C) and the previous study by the authors at 1473 K (1200 °C) has enabled the determination of the effects of temperature on the phase equilibria of the multicomponent multiphase system, including such characteristics as the chemically dissolved copper in slag and Fe/SiO₂ ratio at silica saturation as a function of copper concentration in matte. The new data will be used in the optimization of the thermodynamic database for the copper-containing systems.     

Thermodynamic Activity of Chromium Oxide in CaO‐SiO2‐MgO‐Al2O3‐CrOx Melts

Thermodynamic activities of chromium oxide contained in CaO‐SiO₂‐MgO‐Al₂O₃ melts were measured in the present work using gas‐slag equilibrium technique. The oxygen partial pressure was varied (10^‐3, 10^‐4, 10^‐5 Pa). Gas mixture of CO, CO₂ and Ar were used and investigated at 1803, 1873 and 1923 K. The activities of CrO showed a strong positive deviation from ideality and a decrease with increasing temperature and oxygen partial pressure. A mathematical expression relating the amount of chromium oxide in the slag phase with the activity of Cr in the metal phase based on the present experimental results is presented.     

Manganese Distribution between FeO‐MnO‐SiO2–CaO–MgO‐Al2O3 Slags and Molten Iron

Pretreatment of high manganese hot metal is suggested to produce hot metal suitable for further processing to steel in conventional LD converter and rich manganese slags satisfy the requirements for the production of silicomanganese alloys. Manganese distribution between slag and iron represents the efficiency of manganese oxidation from hot metal. The present study has been done to investigate the effect of temperature, slag basicity and composition of oxidizer mixture on the distribution coefficient of manganese between slag and iron. Ferrous oxide activity was determined in molten synthetic slag mixtures of FeO‐MnO‐SiO₂–CaO–MgO‐Al₂O₃. The investigated slags had chemical compositions similar to either oxidizer mixture or slags expected to result from the treatment of high manganese hot metal. The technique used to measure the ferrous oxide activity in the investigated slag systems was the well established one of gas‐slag‐metal equilibration in which molten slags contained in armco iron crucibles are exposed to a flowing gas mixture with a known oxygen potential until equilibrium has been attained. After equilibration, the final chemical analysis of the slags gave compositions having a particular ferrous oxide activity corresponding to the oxygen potential of the gas mixture. The determined values of ferrous oxide activity were used to calculate the equilibrium distribution of manganese between slag and iron. Higher manganese distribution between slag and iron was found to be obtained by using oxidizer containing high active iron oxide under acidic slag and relatively low temperature of about 1350°C.     

Experimental Investigation of Gas/Slag/Matte/Tridymite Equilibria in the Cu-Fe-O-S-Si System in Controlled Gas Atmospheres: Experimental Results at 1473 K (1200 °C) and <Emphasis Type="Italic">P</Emphasis>(SO<Subscript>2</Subscript>) = 0.25 atm

Experimental studies were undertaken to determine the gas/slag/matte/tridymite equilibria in the Cu-Fe-O-S-Si system at 1473 K (1200 °C), P(SO₂) = 0.25 atm, and a range of P(O₂)’s. The experimental methodology involved high-temperature equilibration using a substrate support technique in controlled gas atmospheres (CO/CO₂/SO₂/Ar), rapid quenching of equilibrium phases, followed by direct measurement of the chemical compositions of the phases with Electron Probe X-ray Microanalysis (EPMA). The experimental data for slag and matte were presented as a function of copper concentration in matte (matte grade). The data provided are essential for the evaluation of the effect of oxygen potential under controlled atmosphere on the matte grade, liquidus composition of slag and chemically dissolved copper in slag. The new data provide important accurate and reliable quantitative foundation for improvement of the thermodynamic databases for copper-containing systems.     

Equilibrium boron distribution between Fe–C–Si–Al melt and boron-bearing slag

HSC 6.1 Chemistry software (Outokumpu) and a simplex–lattice experiment design are employed in thermodynamic modeling of the equilibrium boron distribution between steel containing 0.2% C, 0.35% Si, and 0.028% Al (wt % are used throughout) and CaO–SiO₂–Al₂)₃–8% MgO–4% B₂O₃ slag over a broad range of chemical composition at 1550 and 1600°C. For each temperature, mathematical models (in the form of a reduced third-order polynomial) are obtained for the equilibrium boron distribution between the slag and the molten metal as a function of the slag composition. The results of simulation are presented as graphs of the composition and equilibrium distribution of boron. The slag basicity has considerable influence on the distribution coefficient of boron. For example, increase in slag basicity from 5 to 8 at 1550°C decreases the boron distribution coefficient from 160 to 120 and hence increases the boron content in the metal from 0.021% when L _B = 159 to 0.026% when L _B = 121. In other words, increase in slag basicity favorably affects the reduction of boron. Within the given range of chemical composition, the positive influence of the slag basicity on the reduction of boron may be explained in terms of the phase composition of the slag and the thermodynamics of boron reduction. Increase in metal temperature impairs the reduction of boron. With increase in temperature to 1600°C, the equilibrium distribution coefficient of boron increases by 10, on average. On the diagrams, we see regions of slag composition with 53–58% CaO, 8.5–10.5% SiO₂, and 20–27% Al₂O₃ corresponding to boron distribution coefficients of 140–170 at 1550 and 1600°C. Within those regions, when the initial slag contains 4% B₂O₃, we may expect boron concentrations in the metal of 0.020% when L _B = 168 and 0.023% when L _B = 139.     

Al2O3-ZrO2-Yb2O3 phase diagram. III. Solidus surface and phase equilibria in alloy solidification

The solidus surface for the Al₂O₃-ZrO₂-Yb₂O₃ phase diagram is plotted onto the concentration triangle for the first time. It consists of four isothermal three-phase fields that correspond to two invariant eutectic equilibria and two invariant peritectic equilibria. The solidus surface also includes five ruled surfaces representing the end of crystallization in binary eutectics. The highest solidus temperature in the system is 2710 °C, which is the melting temperature for pure ZrO₂, while the lowest temperature is 1765 °C, which is the melting temperature for the AL+F+Yb₃A₅ ternary eutectic. No ternary compounds or noticeable areas of solid solutions based on components and binary compounds are found in the ternary system. The phase equilibrium diagram and reaction scheme for the equilibrium crystallization of Al₂O₃-ZrO₂-Yb₂O₃ alloys are plotted using data on bounding binary systems and liquidus and solidus surfaces.     

Manganese Capacity and Optical Basicity of Metallurgical Slag

In practice, the concept of slag capacity is used to assess the distribution of elements between condensed phases. In particular, researchers determine the sulfide, phosphate, chromate, and nitride capacity of slags. In the present work, a mathematical model of the manganese capacity is derived. To that end, two equivalent forms of the manganese capacity are derived from the equilibrium constants of the redox reaction of manganese [Mn] + (1/2)O₂ = (MnO). These indices reflect the manganese distribution between the metal and the slag and do not depend on the composition of the metal and the gas phase. One version takes the form C^Mn = K_Mn/γ_(MnO). If we take logarithms and use the known equilibrium constant K_Mn of the redox reaction, we may write logC^Mn = 21122/T–logγ_(MnO)–4.5509. To find the activity coefficient of manganese oxide, equilibrium between hot metal, cast iron, ferrosilicon, ferromanganese, and the corresponding slags is studied experimentally at various temperatures, on circulatory apparatus permitting the study of heterogeneous equilibria involving the gas phase. Using the apparatus, the change in gas volume in the reactions is monitored and automatically recorded and constant pressure is automatically maintained in the system. The attainment of equilibrium is also judged from the constancy of chemical composition of the condensed phases over time. If numerical values of γ_(MnO) are available, they may be used to calculate the manganese capacity of all the slags from the equation already given. For the sake of practical convenience, the manganese capacity is written in terms of the temperature and the optical basicity λ_ed calculated from the electron density known for elements in the periodic table: logC^Mn =–1.866λ_ed + 21049/T–3.131 (R² = 0.997). According to this equation, the manganese capacity depends only on λ_ed and the temperature and may be used for metals and slags of practically any composition.     

Investigation of Liquidus Temperatures and Phase Equilibria of Copper Smelting Slags in the FeO-Fe2O3-SiO2-CaO-MgO-Al2O3 System at PO2 10?8?atm

Copper concentrates and fluxes can contain variable levels of SiO₂, CaO, and MgO in addition to main components Cu, Fe, and S. Metal recovery, slag tapping, and furnace wall integrity all are dependent on phase equilibria and other properties of the phases and are functions of slag composition and operational temperature. Optimal control of the slag chemistry in the copper smelting, therefore, is essential for high recovery and productivity; this, in turn, requires detailed knowledge of the slag phase equilibria. The present work provides new phase equilibrium experimental data in the FeO-Fe₂O₃-SiO₂-CaO-MgO-Al₂O₃ system at oxygen partial pressure of 10⁻⁸ atm within the range of temperatures and compositions directly relevant to copper smelting. For the range of conditions relevant to the Kennecott Utah Copper (South Magna, UT) smelting furnace, it was confirmed experimentally that increasing concentrations of MgO or CaO resulted in significant decreases of the tridymite liquidus temperature and in changes in the position of the tridymite liquidus in the direction of higher silica concentration; in contrast, the spinel liquidus temperatures increase significantly with the increase of MgO or CaO. Olivine and clinopyroxene precipitates appeared at high MgO concentrations in the liquid slag. The liquidus temperature in the spinel primary phase field was expressed as a linear function of 1/(wt pctFe/wt pctSiO₂), wt pctCaO, wt pctMgO, and wt pctAl₂O₃. The positions of each of the liquidus points (wt pctFe)/(wt pctSiO₂) at a fixed temperatures in the tridymite primary phase field were expressed as linear functions of wt pctCaO, wt pctMgO, and wt pctAl₂O₃.     

Prediction of deoxidation and inclusion precipitation in semikilled steel

A generalized thermodynamic calculation procedure for the prediction of oxide inclusion precipitation of desired compositions for Mn-Si-Al deoxidation has been developed. The procedure directly uses established ternary iso-activity diagrams in the MnO-SiO₂-Al₂O₃ system, and does not assume any slag model. The thermodynamic data required for calculation, such as deoxidation constants, interaction coefficients, and activity vs composition relations in the oxide phase, reported in the literature, were examined and the ones considered to be most satisfactory were accepted. On the basis of the computed results for inclusions in the spessartite phase field of the MnO-SiO₂-Al₂O₃ system, discussions have been presented on the applicability of the regular solution model as well as an empirical model reported in the literature. Some predictions for plain carbon steels have also been included.     

Production of ferrosilicomanganese concentrate from tailings slag

The viscosity and electrical conductivity of MnO-SiO₂-CaO-Al₂O₃ slag melts for the largescale smelting of ferrosilicomanganese in RPZ-63 furnaces at PAO Nikol’skii Zavod Ferrosplavov are investigated. Factors responsible for alloy loss are noted, and extraction of the alloy from tailings slag is considered. Instead of pneumatic separation and hydraulic jigging, a new technology is proposed for sorting of the slag and the extraction of metal concentrate, on the basis of electronic sensors and data-processing software.     

Influence of the Al2O3-contents in the ternary-systems (FeO,MnO, CaO,MgO)- Al2O3-SiO2 on the course of the mixture gaps in the ternary space

In accordance with the previous phase equilibria in the ternary systems (FeO, MnO, CaO, MgO)- Al₂O₃-SiO₂ even small Al₂O₃-contents should be sufficient to close the wide mixture gaps in the binary edge systems of FeO-SiO₂, MnO-SiO₂, CaO-SiO₂ and MgO-SiO₂, which are stable in the temperature range from 1700 to 2100-2200°C. Thermodynamically this acceptance is not impossible but rather improb-able. Based on phase-theoretical considerations and the previous results of literature about the course of an unstable mixture gap in the system SiO₂-Al₂O₃ two thermodynamically founded versions were developed as possible, resulting phase equilibria in the SiO₂-rich, but experimentally not exactly examined part of the mentioned ternary-systems, and were added to the experimentally secured sections of these systems unproblematically.     

Calculation of sulfide capacities of multicomponent slags

The Reddy-Blander model for the sulfide capacities of slags has been modified for the case of acid slags and to include A1₂O₃ and TiO₂ as components. The model has been extended to calculatea priori sulfide capacities of multicomponent slags, from a knowledge of the thermodynamic activities of the component oxides, with no adjustable parameters. Agreement with measurements is obtained within experimental uncertainty for binary, ternary, and quinary slags involving the components SiO₂-Al₂O₃-TiO₂-CaO-MgO-FeO-MnO over wide ranges of composition. The oxide activities used in the computations are calculated from a database of model parameters obtained by optimizing thermodynamic and phase equilibrium data for oxide systems. Sulfur has now been included in this database. A computing system with automatic access to this and other databases has been developed to permit the calculation of the sulfur content of slags in multicomponent slag/metal/gas/solid equilibria. Formerly Graduate Student with the Centre for Research in Computational Thermochemistry, Ecole Polytechnique.     

The Effect of CaO on Gas/Slag/Matte/Tridymite Equilibria in Fayalite-Based Copper Smelting Slags at 1473 K (1200 °C) and <Emphasis Type="Italic">P</Emphasis>(SO<Subscript>2</Subscript>) = 0.25 Atm

Fundamental experimental studies have been undertaken to determine the effect of CaO on the equilibria between the gas phase (CO/CO₂/SO₂/Ar) and slag/matte/tridymite phases in the Cu-Fe-O-S-Si-Ca system at 1473 K (1200 °C) and P(SO₂) = 0.25 atm. The experimental methodology developed in the Pyrometallurgy Innovation Centre was used. New experimental data have been obtained for the four-phase equilibria system for fixed concentrations of CaO (up to 4 wt pct) in the slag phase as a function of copper concentration in matte, including the concentrations of dissolved sulfur and copper in slag, and Fe/SiO₂ ratios in slag at tridymite saturation. The new data provided in the present study are of direct relevance to the pyrometallurgical processing of copper and will be used as an input to optimize the thermodynamic database for the copper-containing multi-component multi-phase system.     

Slag oxidation in arc furnaces

In the electrosmelting shop at OAO MMK, the smelting of steel in powerful DSP-180 arc furnaces introduced in 2006 continues to be refined. Industrial research is conducted to establish how the content of the iron oxides FeO and Fe₂O₃ in the slag depends on its total oxidation. As a result, the dependence of FeO/Fe₂O₃ on the total slag oxidation may be described. It is established that the total slag oxidation does not depend on the consumption of injected carbon in frothing the slag. Analysis of the waste gases in the same smelting period shows sharp increase in the content of carbon oxides (especially CO) and the temperature. On account of the poor assimilation of the injected carbon by the melt, much of it is oxidized in the gas phase.     

Phase diagram of the Al2O3 - ZrO2 - Nd2O3 system. III. Solidus surface and phase equilibria in alloy crystallization

A projection has been constructed for the solidus surface in the Al₂O₃ - ZrO₂ - Nd₂O₃ phase diagram on the plane of the concentration triangle, which consists of six isothermal three-phase fields corresponding to two nonvariant equilibria of eutectic type and four nonvariant ones of peritectic type, together with five lineated surfaces for the end of crystallization of the monovariant eutectics. The highest solidus temperature in the system is 2710°C, the melting point of pure ZrO₂ , and the least is 1675°C, the temperature of the ternary eutectic L ? β + F + NA. No ternary phases and no appreciable regions of solid solutions based on their components and the binary compounds have been observed. Data on the adjoining binary systems, liquidus and solidus surfaces allowed for construction of the phase equilibrium diagram together with a reaction scheme for the equilibrium crystallization of alloys in the Al₂O₃ - ZrO₂ - Nd₂O₃ system.     

Effect of coherency stresses on α + α2 phase equilibria in TiAl alloys

Experiments were performed on two TiAl alloys to examine recent theoretical predictions concerning the characteristics of coherent two-phase equilibria in binary alloys. Equilibrium phase compositions in both the coherent and incoherent α + α₂ states at 725 and 800°C were determined by analytical electron microscopy. The results provide experimental evidence for the following predictions: (1) tie-line ends (equilibrium phase compositions) need not coincide with phase boundaries in the temperature-composition phase diagram and (2) the tie-line ends change with alloy composition at constant temperature. The experimental results have been compared with theoretical calculations of α + α₂ phase equilibria incorporating both elastic and ordering effects.     

CAS-OB refining: slag modification with B2O3–CaO and CaF2–CaO

Abstract

Evaluation of CAS-OB refining slags showed that the melting temperature and viscosity were very high and could further increase during the CAS-OB refining process, causing excessive slag to stick to the snorkel with resulting operational problems. To avoid this, B₂O₃–CaO (mass ratio 1 : 1) and CaF₂–CaO (mass ratio 1 : 1) were employed as modifiers added to the slag. The fusibility (melting temperature and viscosity) and desulphurising capacity of modified slag were investigated. Both B₂O₃–CaO and CaF₂–CaO can effectively lower the melting temperature and viscosity of slag. The results of experiments on sulphur partition equilibrium between metal and slag indicate that the sulphur content of metal can be further decreased by the modified slag.