Mineralogical assessment and modelling of the Excalibur pyrometallurgical pilot

Final MSc presentation on the issues encountered during a pyrometallurgy pilot project on FeMn alloy production, such as slag viscosity, refractories corrosion and degassing. Mineralogical (FE-SEM, QEMSCAN, EPMA), chemical (XRF, XRD) and thermodynamic modelling (ThermoCalc, MTData) studies of slag compositions. Final MSc presentation on the issues encountered during a pyrometallurgy pilot project on FeMn alloy production, such as slag viscosity, refractories corrosion and degassing. Mineralogical (FE-SEM, QEMSCAN, EPMA), chemical (XRF, XRD) and thermodynamic modelling (ThermoCalc, MTData) studies of slag compositions. ERAMET owns three plants dedicated to high-carbon ferromanganese alloys. The plants are striving to minimize their environmental footprint, notably by recycling co-products (dust and sludge) captured at various stages in the production process. Two recipes compatible with manganese alloy production have been developed. The first is developed with Eramet Norway; it uses co-products generated at the SAUDA plant to optimize the Group's industrial furnaces. The second recipe was developed with a project funded by EIT Raw Materials, to recover and recycle the manganese contained in the various co-products. The influence of recycling these agglomerated co-products was tested in the pyrometallurgical pilot. This pilot aimed to test these new products' melting performance and suitability for industrial-scale production. Several issues were encountered during this pilot affecting the viability of the process, such as viscosity, refractory corrosion and sulphur content issues. The recipes and the products from the four pyrometallurgy trials were examined to comprehend how their properties affect the pyrometallurgical processes. The mineralogical characterization and thermodynamic modelling enabled the assessment of the matter transformations that occur during the pyrometallurgy processes and the thermochemical and physical variability of the samples. This study demonstrated that all the slags produced during the pilot do not correspond to the fluid manganosite-containing slags produced at the plant. Two major types of slag were identified. A slag with mangano-calcic olivine, omnipresent in all the trials, and a slag that contains manganosite but is highly viscous. The origin of the non-compliance of these slag products is multiple. The first one is an abnormally high reduction in the load and the coke, inducing an insufficient manganese content and a sulphur enrichment. The second is inadequate management of fluxes, resulting in an overabundance of silicon and calcium, increased by constant additions of corrosive fluorite. This implies a chemical composition favouring nucleation and balanced growth of olivine to the detriment of manganosite. The rare manganosite slags formed due to excessive manganese content, allowing massive crystallization at elevated temperatures. Therefore, the mineralogy of the slags produced during EXCALIBUR induced a viscosity too high to obtain a flow. This work enabled an understanding of the problems encountered during the pilot to optimize the ferromanganese alloy production processes.