Loading...
Please wait, while we are loading the content...
Similar Documents
Liquidus surface of FeO-Fe 2 O 3-SiO 2-CaO slag containing Al 2 O 3 , MgO and Cu 2 O at various oxygen partial pressures
| Content Provider | Semantic Scholar |
|---|---|
| Author | Kongoli, Florian |
| Abstract | Liquidus surface of FeO-Fe2O3-SiO2-CaO system is of special importance in general pyrometallurgy. It determines several industrial slags, such as “fayalite” slag, calcium ferrite slag, acidic and basic steel making slags as well as the newly proposed 'ferrous calcium silicate slag', which are all nominally based on this system. Nevertheless the liquidus surface of this system has been studied only in equilibrium with metallic iron or in air and the data at intermediate oxygen partial pressures are almost missing. In reductive smelting processes the above-mentioned slags were based on the liquidus surface of FeO-Fe2O3-SiO2-CaO system in equilibrium with metallic iron. In oxidative matte smelting and in today's modern trends of non-ferrous smelting such as continuous converting or white metal production, the characterization of the liquidus surface of this system at intermediate oxygen partial pressures becomes imperative. Furthermore the effect of minor oxides on the liquidus surface of this system has not been studied at higher oxygen potentials and confusing conclusions are found in literature even for low oxygen potentials. In this work a quantitative description of the liquidus surface of FeO-Fe2O3-SiO2-CaO slag containing Al2O3, MgO and Cu2O is carried out by means of a thermodymamic model at various oxygen partial pressures. Through a new type of easy-to-understand multicomponent phase diagrams it is shown that important differences exist between the liquidus surfaces of this system in reductive and oxidative conditions. It is also shown that minor components can have fundamentally different effects in reductive and oxidative conditions. INTRODUCTION The FeO-Fe2O3-SiO2-CaO system is of special importance in general extractive metallurgy. Many existing industrial slags such as the “fayalite” slag, calcium ferrite slag, acidic and basic steel making slags as well as the newly proposed 'ferrous calcium silicate slag' are nominally based on this system. The liquidus surface of this system determines the operational windows of existing slags as well as the availability of new slags for more advanced technologies. Nevertheless the liquidus surface of this system has been studied only in equilibrium with metallic iron or in air and the data at intermediate oxygen partial pressures are almost missing. However, most of the industrial processes are carried out at intermediate oxygen potentials and not at limiting conditions of iron saturation or in air. For instance, most of the classical copper smelting and converting processes, especially the new modern ones such as continuous converting or white metal production, are carried out at oxygen partial pressures varying from 10 to 10 atm. The liquidus surface of these slags is however not known at these oxygen potentials. Metallurgists seem to have ignored the difference in the oxygen potential and its influence on the slag liquidus surface and continue to empirically use the liquidus surface of the above-mentioned system at iron saturation (oxygen partial pressures from 10 to10 atm) in both reductive and oxidative processes. This is believed to be the reason for the confusion that exists today in the literature and industrial practice about the problem of “magnetite” precipitation which most of the time has been only empirically explained. Most of the industrial slags also contain several amounts of minor oxides, which are introduced through the mineral concentrates, fluxes, dissolved refractories etc. These oxides significantly effect the liquidus surface of FeO-Fe2O3-SiO2-CaO system and the operational windows of the slags. Nevertheless the effect of minor oxides on the liquidus surface of this system has not been studied at intermediate oxygen potentials and more confusing conclusions are found in literature even for low oxygen potentials since the difference between the reductive and oxidative conditions is not taken into account. In previous work (1-3) the liquidus surface of the CaO-FeO-Fe2O3-SiO2-Al2O3-MgO system has been quantified along with the effect of CaO, Al2O3 and MgO as minor oxides at predominantly low oxygen potentials through a new type of multicomponent phase diagrams(1). The purpose of this work is to quantify for the first time the liquidus surface of CaO-FeO-Fe2O3-SiO2 slag containing Al2O3, MgO and Cu2O at various intermediate oxygen potentials. The quantification of the effect of oxygen potential and minor oxides on the slag liquidus surface as well as the clarification of the confusion that exists in today's literature is also envisaged. This work will also serve the purpose of the quantification of the liquidus surface of the newly proposed 'ferrous calcium silicate slag' (4,5). THERMODYNAMIC MODELING A thermodynamic model consists of a set of model equations for the Gibbs energies of all phases in a multicomponent system as a function of temperature, composition and pressure. The parameters of the model equations are obtained by “optimization” in which all available thermodynamic and phase equilibria data in binary and ternary systems of a multicomponent system are evaluated simultaneously to provide a model where the data are self-consistent and obey thermodynamic principles. From these equations all the thermodynamic properties and phase diagrams of simple systems can be reproduced and those of multi-component systems can be predicted. Technically this is realized by storing the model parameters in solution databases, which are then used along with pure component databases by a general Gibbs free energy minimization software that incorporates the model equations to calculate multi-component and multi-phase heterogenous equilibria. This process is described in detail in previous publications (1-11). Thermodynamic modeling is an interesting and useful tool for quantitative characterization of liquidus temperatures and other phase relations as well as thermodynamic properties. It offers three main advantages. Firstly, it considerably reduces the amount of the experimental work needed to characterize the above-mentioned properties over a wide range of compositions and temperatures. Based on the existing experimental data, the model employs thermodynamically and structurally correct interpolations and extrapolations to predict the liquidus temperature on several regions of composition and temperature where the data do not exist. Only a few new experimental data may be needed to verify and calibrate the model predictions in the multicomponent systems. As a consequence, the time and overall cost associated with a quantitative characterization of the liquidus surface in a wide range of compositions is decreased. Secondly, thermodynamic modeling can overcome some difficulties occurred during the measurements of the slag liquidus of the multicomponent slags. Usually these slags have steeply sloping surface whose placement depends on the stable primary phase. Not only is the appearance of the first solid phase difficult to detect, but in addition the liquidus temperature is so sensitive to compositional changes that normal errors in chemical analyses will result in data scatter. This explains the fact that the experimental liquidus data are usually quite scattered and difficult to correlate using simple empirical model. In this work several new private models for the liquid and solid phases in the CaOFeO-Fe2O3-SiO2-Al2O3-MgO system were developed by the authors. In addition, old private and FACT models were modified by the authors in the light of the new data (12) (to be published elsewhere). The modified quasichemical model (10) was used for the liquid slag. The compound energy formalism (14) as well as other polynomial models were used for modeling of solid phases. Private software(12) and FACT system (13) were used for the construction of the diagrams and the modeling process. The predictions of the model were first verified against all available experimental data related specifically to the oxygen potential and minor oxides. This process was followed by the construction of simple polythermal projection and isothermal diagrams in order to quantify their effect on the liquidus temperature of the CaO-FeO-Fe2O3-SiO2 slag and to quantitatively describe the important differences that exist between oxidative and reductive conditions. EFFECT OF OXYGEN ON THE LIQUIDUS SURFACE OF FeO-Fe2O3-SiO2-CaO SLAG Experimental Verification of Model Predictions Figures 1-3 show the liquidus isotherms in CaO-FeO-Fe2O3-SiO2 system at 1300 C and at several oxygen partial pressures (10, 10 and 10 atm) according to model predictions and the recent experimental data of Tsukihashi (15). The data of Takeda et al.(16) and Takeda (17) in the limiting ternary systems are also shown. The agreement between the model and the experimental data is within experimental error limits. A disagreement though exist between the model and the data of Tsukihashi(15) at iron oxide-rich corner at oxygen partial pressure of 10 atm. However, taking into account the fact that the model agrees with the data of Takeda et al.(16) and Takeda(17) in the limiting ternary systems and the wellknown fact that an increase in the oxygen partial pressure stabilizes magnetite, it can be said that, at this particular region, the experimental data may not be correct. Some other experimental data on the liquid regions of this system at 1300C and oxygen partial pressure of 10atm has been reported by Shigaki et al.(18). These data do not agree with the data of Tsukihashi (15) in the quaternary system and those of Takeda et al. (16) and Takeda (17) on the limiting ternary systems, which, as shown above, satisfactorily agree with the model. It should be noted that the above-mentioned model predictions are based exclusively on the binary and ternary sub-systems and the predicted diagrams at these oxygen partial pressures were constructed before the recent data of Tsukihashi(15) became available. Polythermal Projecti |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://www.pyrometallurgy.co.za/MoltenSlags2000/pdfs/023.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
