Ferromanganese and silicomanganese are ferrous alloys used primarily in steel production. They are composed of iron and manganese, and silica is also present in silicomanganese. TEMCO is a subsidiary of Samancor, and operate a ferro- and silicomanganese smelter located at Bell Bay, Tasmania, which uses electric arc furnaces to reduce manganese oxide ore to manganese metal to form the two different alloys.
The furnaces use graphite bricks as refractory lining, which must be periodically replaced. The aim of this thesis was to develop a test to allow examination the processes occurring at the slag/refractory boundary, and to identify any phases which are present. A test has been developed which allows a refractory crucible containing slag and alloy to be heated to specific temperatures for a controlled lengths of time. It allows the phases formed at the interface to be identified after quenching of the sample.
The test was developed over a series of experiments, which used ferromanganese slags and graphite refractories supplied by TEMCO. TEMCO has found a solid layer forms at the refractory wall in the industrial furnaces. Thermodynamic calculations indicated that this could possibly be due to a reaction between slag and the carbon refractory forming manganese carbide. Experiments were conducted which used a tube furnace to heat slag/alloy-containing crucibles to two temperatures, one above and one below the temperature at which the carbide forming reaction was predicted to take place.
A number of samples were also obtained from TEMCO which were taken from the furnace during a dig-out. Analysis of these samples allowed the comparison of results from the laboratory tests to the processes actually occurring within the industrial furnace.
Analysis of the crucibles from the laboratory tests and the samples from TEMCO were analysed using optical microscopy, electron probe microanalysis and X-ray diffraction analysis. Phases identified forming in the laboratory crucibles included a glassy slag phase, a metallic phase forming at the slag/crucible interface, possibly ferromanganese alloy or manganese carbide, gehlenite, galaxite and manganosite.
Analysis of the TEMCO samples showed similar phases to those in the crucibles, however in different proportions. Results indicated that a solid manganosite and galaxite exists at the furnace wall due to cooler temperatures around the refractory. These phases exist as grains in a tephroite or kalsilite matrix.
Close to the furnace wall, a phase K2O.Al2O3.SiO2 is prevalent, indicating potassium is concentrating at the refractory surface.
The presence of solids at the industrial refractory/slag interface indicates that isothermal tests conducted around the liquidus of the slag may not give an accurate representation of processes in the furnace. Future tests should be conducted at temperatures closer to those that exist at the furnace wall.
Other recommendations include:
• An investigation into the affects of potassium on graphite refractory, as results showed a high concentration of potassium near the refactory wall,
• Tests be conducted using alternative refractory types to allow comparison between the performance of different refractories,
• Tests be conducted which monitor off-gases to indicate whether any reactions are taking place within the furnace.
Project and risk management have been undertaken, using a Gantt chart to schedule the project and Workplace Risk Analysis and Control, Job Safety Analysis and Material Safety Data Sheets to predict risks to the thesis and personal safety.