On the surface quality of continuously cast steels and phosphor bronzes

Sammanfattning: This thesis work concerns about the importance of the cast surfaces, surface phenomenon such as the formation of the oscillation marks and exudation and related defects including cracks and segregation that happened during the continuous casting. All of the investigated materials were collected during the plant trials while an in-depth analysis on these materials was performed at the laboratory scale with certain explanations supported by the schematic and theoretical models. The work consists on different material classes such as steels and phosphor bronzes with a focus on the surface defects and their improvements.Metallographic investigation was performed at the surfaces of three steel grades, including Incoloy alloy 825, 4HSL64 and 82700. Incoloy alloy 825 solidifies primarily to austenite while other steel grades show peritectic reaction/transformation. The samples from Incoloy alloy 825 and 4HSL64 were taken from the surface of a continuously cast bloom, while 82700 was produced as continuously casted slab. Samples were collected from the heat under casting conditions which are normally used to produce the corresponding steel grades. Sampling was done at different locations across the width of the strand. The microscopic analysis included all of the marks encompassed in the slice of the materials with a length of 250 mm along the casting direction. The characteristics of the marks were investigated by importing and serially arranging all the micrograph to the drafting software, AutoCAD, with high precision. The metallographic study was combined with macro/micro examinations of oscillation marks and segregation analysis of Cr, Mn, Mo, Ni and Si by microprobe analysis.The results showed that OMs have widely different characteristics, such as mark type, pitch, span, depth, segregation at the mark bottom and accompanying microstructure. However, the average mark pitch and mark span have found to be given by the simple analytical relations. Furthermore, the mark pitch can vary considerably even for similar casting conditions, leading to different conditions for the marks formation in relation to the mold’s cyclic movement. The variation in the observed hook lengths and shape could be attributed to the varied mark pitch. Moreover, the type of the marks found to be indistinguishable while looking at the cast surface. Nevertheless, the wider marks were deeper. Based upon the observations, a mechanism for the formation of the marks was proposed where the direction of the mold movement played an important role to define the mark type. The deviation in the characteristics of OMs was ascribed to the unsynchronized mold movement and development of the meniscus. The mechanism also explained the formation of the observed defects such as cracks and segregation at the bottom of the marks.However, for the case of peritectic steel grades, the formation of the defects were found to be less relevant to the formation of the oscillation marks. The role and mode of the peritectic reaction/transformation is found to be dependent on the composition of the alloy, resulting in different types of surface cracks. The effect of the local variation in the cooling rate on the formation of the different types of cracks present in each steel grade, such as is due to the formation of the oscillation marks, is demonstrated. The enhanced severity of the surface and internal oxidation, both of which dependent on the alloy composition and consequent peritectic reaction, is highlighted. Experimental and theoretical studies show that different types of surface cracks can occur in peritectic steels depending upon the alloy composition and cooling rate, both of which define the fraction of the remaining liquid upon completion of the peritectic reaction/transformation. Possible solutions for minimizing the observed defects by optimizing the mold conditions are suggested.Alloys with tin content 2-8 wt-% were casted during plant trials of the vertical thin strip continuously cast phosphor bronzes. Surface investigation of the obtained samples showed that the formation of oscillation mark did not affect the cast negatively. Instead, the spread of the tin rich liquid at the chill surface caused by the formation of flow channels underneath the chill surface found to cause the sticking of the casting to the mold walls. The severity of the exudation of the liquid increased with the increase in the tin content. Precipitation of the secondary phases has also been observed under some casting conditions. The macro segregation profile along the solidification thickness predicts a strong casting parameter sensitive inverse segregation. The simulation results show high compressive stresses at the surface of the cast during solidification. The observed flow channels and calculated thermal stresses shows the possibility of the pressure driven flow of tin rich liquid towards the chill surface during solidification. It is suggested to cast phosphor bronzes with increased liquid pool depth which eventually reduces the temperature gradients along the casting direction, thus the spread of the exudate layer.In order to facilitate the theoretical analysis which could be capable of explaining the suggested phenomenon in the thesis for steel and phosphor bronzes, a reduced model is developed which required lesser computational resources with lesser convergence problems. The comparison of the reduced and full model showed a reasonable agreement. The model calculates the phase change interfaces during the solidification explicitly with stable gradient which unlike the full problem are very useful for further computations. The model is further extended to compute the macrosegregation profile. Systematic asymptotic analysis of the steady-state two-dimensional mass, momentum, heat and solute conservation equations in terms of the shrinkage parameter indicates that the overall problem can be reduced to a hierarchy of decoupled problems: a leading-order problem that is non-linear, and a sequence of linear problems, with the actual macrosegregation of the solute then being determined by means of one-dimensional quadrature.