Thesis (PhD.(Metallurgy)--University of Pretoria, 2007. / Includes bibliographical references.

The automotive industry, has been innovating in the field of materials development in order to meet the demand for lower emissions, improved passenger safety and performance. Despite various attempts of introducing other lightweight materials (Al, Mg or polymers) in car manufacturing, steel has remained as the material of choice till date due to its excellent adaptability to systematic upgradation and optimization in its design and processing. One of the outcomes is the development of second generation high Mn TWin Induced Plasticity (TWIP) steels with excellent strength-ductility balance suitable for automotive applications. Cost effective high performance TWIP steel design is mostly based on its alloy design and advanced up and down stream processing methods (thermomechanical controlled processing (TMCP)) which can help achieve suitable microstructure to meet the property requirements. It has been observed that grain boundary migration (GBM) in austenite during high temperature TMCP stage dictates grain growth to control the final microstructure. This research work initially investigates the grain growth in Fe-30%Mn steel within a temperature regime of 1000-1200°C. Compared to conventional low Mn steel, austenite boundary mobility in Fe-30%Mn was found to be 1-2 orders of magnitude smaller. Atom probe tomography results showed no Mn segregation at austenite high angle grain boundaries (γ-HAGB) which rules out the effect of Mn solute drag on growth kinetics in Fe-30%Mn steels. Grain boundary character distribution (GBCD) study showed that the sample consists of two different population of grain boundaries. 50% of the grain boundaries are random HAGBs with high mobility. Remaining 50% are special in nature which introduce low mobility boundary/boundary segments in the global boundary network. The special boundaries are mostly in the form of Σ3 CSL boundaries or its variants like Σ9, Σ 27. These boundary/ boundary segments were introduced by the formation of annealing twins and their interactions with the random HAGBs. An attempt to investigate the effect of Mn on growth kinetics at 1200°C showed that Mn slows down growth kinetics up to 15 wt% predominantly by the formation of annealing twins. A qualitative study of the microstructures showed that as Mn concentration is increased from 1% to 15%, the annealing twin density increases resulting in Σ3 frequency to be 30%. The increased twinning frequency is attributed to the effect of Mn on lowering the stacking fault energy (SFE). Annealing twins, belonging to Σ3 CSL family, intersect the HAGBs resulting into twin induced boundary segments which possess very low mobility. In the light of this idea, slow grain growth in high Mn steel was attributed to the population of low mobility boundaries. The proposed ‘twin inhibited grain growth’ model clearly points to the low mobility boundary/boundary segments to be the rate controlling factor during grain growth in high Mn steels. The effect of carbon on grain growth in Fe-30%Mn steel showed that the presence of carbon makes the growth kinetics faster by a factor of 4 and 6 at 1200°C and 1100°C respectively. Although, atom probe tomography results indicated that in presence of carbon, Mn segregation takes place at γ-HAGBs in Fe-30%Mn steel, solute drag does not appear to play a role as it was seen that with increase in Mn content beyond 1%, the solute effect of Mn in slowing down HAGB migration becomes weak. Also, abovementioned higher mobility values are obtained from the growth kinetics of Fe-30Mn-0.5C. This once again highlights the fact that effect of Mn in slowing down grain growth is due to the low mobility of twin/twin related boundaries or boundary segments. Controlling grain growth has been commonly proposed to be accomplished through small addition (<0.1%) of microalloying elements (Nb, V and Ti) which can slow down GBM at high temperature by solute drag and at low temperature by precipitate pinning (Zener drag). This research work has also experimentally quantified the solute drag of Nb in a series of Fe- 30%Mn steels. Grain boundary mobility was estimated for various temperatures and niobium contents. An attempt was made to calculate the grain boundary mobility in presence of niobium using Cahn’s solute drag model. This calculated mobility, when used in the proposed ‘twin inhibited grain growth’ model, the predicted growth kinetics which showed very good fit with the experimentally obtained growth kinetics in case of Fe-30Mn-0.03Nb and Fe-30Mn-0.05Nb steels at 1100°C. The effect of Nb solute drag, thus captured using Cahn’s model, was shown to be slowing down only the HAGB migration in the microstructure, whilst the special boundary mobility was not affected by solute Nb. Another attempt was made through grain boundary engineering (GBE) to control grain growth in Fe-30Mn-0.5C steel. Using different TMCP schemes, GBCD was modified to produce maximum frequency of special boundary. Preliminary studies on grain growth of single step-grain boundary engineered samples did show a significant lowering of grain size compared to a no-GBE sample after grain growth. However, the effect of iterative GBE didn’t show any significant effect in controlling grain growth in spite of the fact that it increased Σ3 frequency to 64%. This probably indicates that the effect of GBE on grain growth by the formation of annealing twins/special low mobility boundaries is a complicated process which might involve twin/special boundary morphology, annihilation kinetics and formation of grain clusters in the microstructure other than the formation of immobile special triple junctions through the intersection of twins/special boundaries with the random HAGBs. / Thesis / Doctor of Philosophy (PhD)