Reduction Rates of SiMn Slags from Various Raw Materials

Abstract

The kinetic information in the SiMn (Silicomanganese) process from melting of raw materials to production of metal was investigated in this work. This is of interest for the SiMn production industries where the effect of raw materials in the furnace is not well known and is assumed to give impact to the metal producing rates. The doctoral work comprises experimental investigation of industrial and synthetic SiMn charges, where a TGA (ThermoGravimetric Analyzer) under CO atmospheric pressure was used to simulate the industrial SiMn furnace. The kinetic information was observed and confirmed by considering four subsequent research topics. First, the melting behavior of raw materials with particle sizes between 4 and 20 mm in SiMn charges was initially investigated mainly between 1200 and 1400 °C. Cross-section images and micro-analyses of two different charge compositions with and without HC (High-Carbon) FeMn (Ferromanganese) slag as raw material were compared to observe the slag forming temperatures. The results from the cross-section images proved that the melting of charges materials and completion of liquid slag occur relatively fast and at low temperatures between 1200 and 1400 °C regardless of particle size and of using HC FeMn slag as raw material. In fact, the contact between manganese sources and quartz was the most important factor during the formation of liquid slag, while the individual melting temperatures of each raw material were of less relation. From the micro-analyses, it was observed that the formation of liquid SiMn slag complies to the binary MnO-SiO2 system regardless of the charge composition. The slag phases were mainly composed of the two manganese silicates, Mn2SiO4 and MnSiO3, which were solidified from liquid SiMn slag. Second, the reduction behavior of SiMn charges were investigated by comparing FeMn charges as reference between 1200 and 1600 °C. Charges with two different particle sizes, 0.6 − 1.6 and 4.0 − 6.3 mm, were also compared to observe the effect of particle size on the reduction rate. It was observed that the reduction of SiMn charges had occurred in two stages, while the reduction of FeMn charges was progressive. The two-stage reduction of SiMn charges involved a slow reduction followed by a rapid reduction, where the dividing temperature was approximately 1500 °C. The effect of particle size for reduction was only observed with FeMn charges, where faster reduction was observed from charge with smaller particle sizes. SiMn charges only contained liquid slag regardless of the particle sizes, which was observed from the experiments of the melting behavior of raw materials. Completion of liquid slag before the second reduction stage had nullified the effect of particle sizes on the reduction rate. Third, the kinetic information of MnO and SiO2 reduction between 1500 and 1650 °C were obtained by using the following rate models:

The chemical reactions of MnO and SiO2 reduction were assumed to be the rate-determining steps, where the rate and kinetic parameters were calculated by investigating three different SiMn charges: Charges “As” (Assmang ore + quartz + coke), “As/HCS” (Assmang ore + quartz + HC FeMn slag + coke) and “HCS” (quartz + HC FeMn slag + coke), where the manganese bearing materials are Assmang ore (As) and HC FeMn slag (HCS). It was initially observed that SiMn charges containing HC FeMn slag as raw material had faster and higher reduction rates than charges without HC FeMn slag. The use of HC FeMn slag had the apparent effect of the enhanced reduction rates. From the comparison of the rate and kinetic parameters, it was observed that the reduction rate of MnO was considerably more influenced by the slag properties from the rate constants rather than the driving force (aMnO − aMn/KMnO). The comparison also indicated that the effect of sulfur as impurity element in the charge was superior than the effect of slag viscosity or amount of iron for enhanced reduction rates. In addition, the two rate models were applicable to describe the changing amount of MnO and SiO2 in various SiMn slags between 1200 and 1650 °C of different reduction degrees. Finally, the effect of sulfur for enhanced reduction rates was confirmed by experiments with synthetic SiMn charges. A controlled slag system, MnO-SiO2-CaO, with different amount of sulfur between 0 and 0.9 wt% was studied between 1500 and 1650 °C. Addition of sulfur into the charge has significantly enhanced the reduction rates of MnO and SiO2, where the threshold amount of sulfur was approximately 0.3 wt%. Also, it was observed that sulfur does not behaves as a catalyst for MnO reduction but was assumed to influence the reduction rate between slag and dissolved carbon in the metal phase by increasing the wetting between the two phases. Similar observation of the effect of sulfur from iron and steel production studies were compared with this work, where the effect of sulfur was further discussed. In addition, the reduction paths of MnO and SiO2 in the MnO-SiO2-CaO system between 1200 and 1650 °C were construed by using the two rate models and experimental measurements. It was observed that the initial direction of the reduction was determined by the SiO2/CaO ratio, but the reduction degree was determined by the amount of sulfur.