Degree Name

Doctor of Philosophy


Department of Materials Engineering


Experiments were carried out to study the grain growth during reheating and phase transformation behaviour during continuous cooling of austenite in five microalloyed steels. Two of these steels were supplied in as-cast slab as well as in controlled rolled plate conditions. These steels contained similar levels of C (~ 0.07%), Ti (~ 0.013%) and Mo (~ 0.2%) but had different levels of Mn (1.1 to 1.7 %) and Nb (0.03% to 0.06%) contents. The principal aim of this work was to simulate as closely as possible the controlled rolling schedule used in industry for the production of plates in laboratory. Laboratory process simulation was carried out using quench and deformation dilatometry. Continuous cooling transformation behaviour of the undeformed as well as thermomechanically processed austenite was investigated from the dilatometric data, quantitative optical metallography and microhardness tests. Continuous cooling transformation (CCT) diagrams of the undeformed and thermomechanically processed steels were constructed. Effects of thermomechanical processing (TMP), accelerated cooling and Mn and Nb contents on austenite transformation critical temperature (Ar3), phase transformation kinetics, CCT diagrams, grain size and hardness were investigated.

The experimental work on the grain growth of austenite during reheating involved mainly the determination of Grain Coarsening Temperatures (GCT) for various microalloyed steels. It was found that the conventional method of detecting the GCT from the mean grain size v/s reheating temperature plot was inadequate in locating the abnormal grain growth in some microalloyed steels. An alternative quantitative metallographic method which consists of plotting the mean size of the largest grain function of reheating temperature was therefore developed. The new method was found to be more effective and sensitive in detecting the abnormal grain growth in all microalloyed steels studied. The factors in addition to precipitate dissolution and coarsening which influence the GCT of microalloyed steels were investigated. It was found that a high heterogeneity ratio (Z) present in the initial austenite grain structure (prior to coarsening) can result in lowering the GCT significantly. The reasons why different steels exhibit different values of heterogeneity ratios were also investigated. was found that microstructural features before reheating influences the α→γ transformation which results in different values of Z in different steels. In slab steels, microstructure containing coarse ferrite grains and a small volume fraction of acicular ferrite + pearlite resulted in a high value of Z. In case of plate steels, microstructure consisting of heterogeneous distribution of ferrite grain sizes and a non-uniform distribution of pearlite (banded microstructure) led to the generation of high Z in austenite grain sizes. Grain growth behaviour of plate steels was found to be significantly different than that of slab steels of the same composition. Plate steels exhibited two stages of accelerated grain growth compared to only one stage of accelerated grain growth found in slab steels. Also, abnormal grain growth in plate steels was found to be not as significant as that in slab steels.

Empirical mathematical models to predict the grain growth of austenite were critically reviewed. It was found that, for the same starting conditions, predictions from these models differ from one other considerably. Furthermore, none of the models could adequately describe the grain growth behaviour, particularly at higher temperatures, of as-cast slabs. Basic principles of grain growth were therefore revisited in an attempt to develop a mathematical model which can account for abnormal grain growth. A model was developed which also accounts for the effect of initial grain size on the process of grain growth. Supporting mathematical models for the prediction of particle coarsening and dissolution as a function time, temperature and composition were developed. It was found that predictions based on this model were in close agreement with the experimental grain growth data.

In the second part, an effort was made to characterise the TMP used in the industry for the production of controlled rolled plates and then to simulate this process as closely possible in the laboratory using quench and deformation dilatometry. Effects of TMP, accelerated cooling and alloying elements (Nb and Mn) on Ar3 temperature, phase transformation kinetics, grain size, CCT diagrams and microhardness were investigated. It was found that the TMP accelerated the onset of γ/α transformation (γ transformation start temperature, Ar3, was raised). However, the progress of γ/α transformation was retarded considerably in the deformed samples. TMP lowered the hardenability of austenite considerably, thus lowering the hardness of the deformed samples. TMP had pronounced effects on the CCT diagrams which include:
1. γ/α transformation was raised to higher temperatures,
2. γ/α transformation nose was shifted towards faster cooling rates; and
3. γ/α transformation occurred over a wider temperature range.
TMP also refined the ferrite grain size significantly.

Increase in cooling rate lowered the Ar3 significantly and also accelerated the progress of γ transformation. Increased cooling rate allowed the transformation of γ to nonequilibrium phases such as bainite and martensite and thus increased the hardness of undeformed as well as deformed samples. Increased cooling rate was found to have marginal effect on the refinement of ferrite grain size. For similar levels of other alloying elements, an increase in N b content was found to lower the Ar3 of undeformed samples at a rate of about 10 °C/ 0.01% dissolved Nb. However, in deformed samples, increase in Nb content actually raised the Ar3 temperature. This effect was considered to be due to two main reasons: firstly, the strain induced precipitation of Nb in austenite which would act as nucleation sites for phase transformation and secondly, due to the scavenging of Nb during precipitation which decreases the hardenability of austenite. Higher Nb content retarded the progress of γ transformation of undeformed as well as deformed austenite. Increase in Nb content increased the hardness of undeformed as well as deformed samples and refined ferrite grain size marginally. Nb had no significant influence on the location of phase transformation noses in the CCT diagrams.

For similar levels of other alloying elements, an increase in Mn content was found to lower the Ar3 of undeformed samples. In contrast with Nb, increased Mn content lowered the Ar3 of deformed samples as well, suggesting that Mn is not scavenged during the strain induced precipitation of Nb and remains dissolved in the austenite which increases its hardenability. Increase in Mn content retarded the progress of transformation of undeformed as well as deformed samples. Increased Mn content increased the hardness of both undeformed and deformed samples significantly. Mn also refined ferrite grain size. Bs temperature was significantly lowered due to a higher Mn content. Increase in Mn content pushed the polygonal ferrite nose in CCT diagrams to slower cooling rates which indicates increased hardenability of austenite and thus counters the effect of TMP which decreases the hardenability.