Degree Name

Doctor of Philosophy


School of Mechanical, Materials and Mechatronic Engineering


Solidification microstructure is a defining link between production techniques and the resulting mechanical properties of cast metals and in particular steel. The microstructural scales of these structures control the segregation profiles of solute elements in the interdendritic regions, thus determining the mechanical properties of the cast structure. Hence, understanding the various aspects of solidification is essential in the quest for the fabrication of better products.

The microstructure achieved during solidification of alloys is a result of morphological evolution of the solid/liquid interface. In the past, it was shown that some of the contributing factors which influence the solid/liquid interface morphology are: temperature gradient, solidification rate and alloy composition. Real-time solidification studies in iron-based alloys are particularly difficult due to their occurrence at high temperatures. Consequently, real-time observations of solid/liquid interface morphologies are mostly limited to the study of transparent organic materials or alloys with relatively low melting points, using Bridgman-type furnaces and directional solidification techniques.

In Chapter 2, a Laser Scanning Confocal Microscopy (LSCM) was used to experimentally study the microstructural formations during solidification of steel. The major objective of this chapter was to investigate the feasibility of the LSCM and the concentric solidification technique as an alternative to the traditional directional solidification methods in Bridgman-type furnaces. Experimental observations were carried out on a low-carbon, low-alloyed steel. The effect of varying amounts of copper additions on the interface instability was studied and subsequently discussed.

The LSCM provided the possibility of real-time observations at temperatures as high as 1600◦C. The conventional LSCM’s furnace was equipped with a controller that cooled the specimens at controlled rates of up to 100◦C/min. Cooling rates of up to 2500◦C/min were also achieved by momentarily turning off the furnace power for short periods of time. By doing so, the microscope chamber cooled at rates which were measured subsequently by acquiring data from the thermocouples. In contrast to the Bridgman apparatus, where controlled temperature gradients can be induced, the temperature gradient cannot be externally controlled in the LSCM apparatus. The technique used in this study therefore reproduced a non-steady-state growth condition. In-situ observations provided the opportunity to study the planar to cellular interface transitions in steel and revealed some unique phenomena associated with this transition.

In Chapter 3, MICRESS multi phase-field method was used to simulate the pattern formation during solidification of the low-carbon low-alloyed base steel. The model was linked with in-situ measurements of solid/liquid interfacial energy in the LSCM, to obtain a more reliable prediction of interface stability during solidification. Once a “Standard model” achieved, the temperature gradient and cooling rate in the model were extended beyond the experimental limiting conditions of the LSCM. A stable planar solid/liquid interface was initially produced in the model. This stable solid/liquid interface was then subjected to temperature gradients of up to 100◦C/mm and cooling rates as high as 1000◦C/min. Solute segregation at different stages of interface transition also modeled using the solute segregation component of the MICRESS code. Discussed, are the solute segregation profile (1) at the solid/liquid interface at the onset of planar to cellular transition, (2) ahead of the dendrite tips and (3) in the inter cellular/dendritic liquid. The segregation model revealed the conditions of solute segregation as a criterion for the prediction of the planar to cellular to dendritic interface transition, the dendrite ternary-arm formation and the dendritic to seaweed transformation.

In Chapter 4, the “Standard model” of Chapter 3 was further extended in order to simulate the solidification microstructure of a low-carbon steel strip produced in a twin-roll caster. The effect of copper on the solidification microstructure and the segregation of copper were subsequently investigated in the model. Study of copper segregation in the steel strip is of scientific as well as industrial significance because copper is introduced into the melt by copper containing scrap which is typically used as feedstock in the twin-roll strip casters.