Inclusion removal is critical for the production of clean steel. A better understanding of removal processes require knowledge of the effect of process parameters on dissolution kinetics. The present research focuses on the kinetics of calcium aluminate inclusion dissolution in relevant steelmaking slags that contain CaO, Al2O3 and SiO2. In-situ observation of inclusion dissolution in slag is conducted using a high temperature, confocal scanning laser microscope (HT-CSLM). The particles used in this experimental work are produced in the laboratory and the production technique is explained in detail. The change in particle size is recorded with time and the effects of temperature, slag composition and inclusion morphology are investigated. The images are extracted from video and they are analysed to record the change in equivalent radius of a single particle during the dissolution process. The original and normalized dissolution data is used to determine the dissolution mechanism and to improve existing dissolution models. It has been found that an increase in temperature increases the dissolution rate. At 1550°C and 1600°C, there is no product layer formation at the slag-inclusion interface and so, the dissolution process is faster. Slag composition shows a significant influence on the dissolution kinetics due to differences in the dissolution driving force and viscosity. Additionally, the dissolution rate depends on the morphology of inclusion as available reaction sites vary significantly. Rate limiting steps are discussed based on the shrinking core model and diffusion in stagnant fluid model. It is shown that the rate limiting step for dissolution is the diffusion v through a product layer at 1500°C whereas it is mass transfer in slag at 1550°C and 1600°C. The diffusion coefficient of alumina is obtained by applying a one-dimension diffusion model. The calculated results varied between 5.5×10-11 and 2.6×10-10 m2/s depending on experimental conditions. Slag viscosity was found to be an important parameter for the modelling of the dissolution process. A modification to the correlation between the correction coefficient and slag viscosity was proposed. This modification improved the prediction of the dissolution path for calcium aluminate and alumina inclusions in steelmaking slags. This novel study provides an understanding of dissolution mechanisms and it offers data on the dissolution rate of CA2 inclusions in the slags related to the process of steelmaking. The results from this work can be used by steelmakers to aid in process design. / Thesis / Master of Applied Science (MASc) / The present work is a pioneer study on the dissolution of calcium aluminate particles in liquid oxide mixtures using the unique real-time observation approach. Experiments were conducted to provide a better understanding of the effects of various steelmaking conditions on inclusion removal during the refinement of liquid steel. An existing dissolution model is further refined by introducing an additional parameter that is correlated to the properties of oxide mixtures. It has been found that the dissolution model can be applied not only to calcium aluminate inclusions but also to alumina inclusions. Hence, the approach proves the potential university nature of the dissolution model. A clear understanding of the dissolution kinetics of inclusions helps to optimize the current steelmaking routes and enhance the removability of inclusions. Steel with a minimum amount of inclusions has better properties from all aspects, which improves its applicability in all fields.

The grades of steel used for automotive bodies are interstitial free steel grades and titanium stabilized ultra-low carbon steel grades. During the manufacturing of these grades, the addition of titanium in liquid steel is achieved in the steel refining units and may cause processing problems. Titanium reacts with the dissolved aluminum and oxygen to form complex solid aluminum titanate type micro-particles (inclusions). During the flow of titanium alloyed steel grades containing solid inclusions (such as aluminum titanate and alumina type inclusions), the clog accompanied by steel skull can be formed at the submerged entry nozzle between the tundish and the mold. To reduce the effects of aluminum titanate type inclusions, they can be either modified or removed. The current study focused on the removal of Al2O3, TiO2, and Al2TiO5 inclusions by dissolving them in slag in the temperature range of 1430 – 1600 °C using a high-temperature confocal scanning laser microscope. In this technique, a single particle (inclusion) is placed on the surface of a solid slag, and the inclusion-slag system is heated to steelmaking temperatures. The dynamic changes in inclusion size are measured to determine dissolution kinetics and mechanism. This work has developed a complex oxide particle synthesis technique and provides the first-ever kinetic data for removal of aluminum titanate inclusions into steelmaking slags. It is found that Al2O3 inclusions have a slower dissolution rate than that of Al2TiO5 inclusions followed by TiO2 inclusions. The rate-controlling steps are investigated using a shrinking core model. It is shown that the rate-controlling step for dissolution of both Al2O3 and Al2TiO5 inclusion types is the mass transfer of alumina. Evidence in support of this finding is the particle-slag interface characterization by line scan analysis and calculated diffusivity values being inversely proportional to the viscosity of slag. The dissolution path of aluminum titanate is proposed in the following steps. First, aluminum titanate dissociates into alumina, titanium oxide and oxygen while slag penetrates through the particle. In the next step, the alumina and titanium oxide dissolves in slag, and the oxygen leaves the system. The existence of gas bubbles enhances the overall rate of Al2TiO5 dissolution. The current work establishes a detailed understanding of the dissolution of Al-Ti-O type inclusions in steelmaking slags. This knowledge will inform steelmakers on which inclusions of different chemistry can be removed preferably and develop strategies on better slag design to produce superior quality steel with reduced operational downtime. / Thesis / Doctor of Philosophy (PhD)

Optical imaging techniques have improved much over the last fifty years since the invention of the laser. With a high brightness source many imaging applications which were once inaccessible to researchers have now become a reality. Among these techniques, the most beneficial one is the use of lasers for both wide-field and confocal imaging systems. The aim of this study was to design a laser imaging system based on the concept of laser scanning confocal microscopy. Specifically the optical system was based on optical fibers allowing the user to image remote areas such as the inner surface of rifled gun barrels and/or pipes with a high degree of precision (+/- 0.01 mm). In order to build such a system, initially the theoretical foundation for a confocal as well as a wide-field imaging system was analyzed. Using this basis a free-space optical confocal system was built and analyzed. The measurements support the fact that both the objective numerical aperture and pinhole size play an important role in the radial and axial resolution of the system as well as the quality of the images obtained. To begin construction of a confocal, optical-fiber based imaging system first an all fiber wide-field imaging system was designed and tested at a working wavelength of 1550 nm. Then an all fiber confocal system was designed at a working wavelength of 808 nm. In both cases results showed that while lateral resolution was adequate, axial resolution suffered since it was found that the design of the optical system needs to take into account under-filling of the objective lens, a result common with the use of laser beams whose divergence is not at all like that of a point source. The work done here will aid technology that will be used in the elimination process of faulty rifling fabrication in defense industry. The reason why the confocal technique is preferred to the conventional wide-field one is the need for better resolution in all directions. Theoretical concepts and mathematical background are discussed as well as the experimental results and the practical advantages of such a system.