Kinetics

The rate of the bainite reaction needs to be considered in terms of a number of distinct events. A sub-unit nucleates at an austenite grain boundary and lengthens at a certain rate before its growth is stifled by plastic deformation within the austenite. New sub-units then nucleate at its tip, and the sheaf structure develops as this process continues. The overall lengthening rate of a sheaf is therefore smaller than that of an individual sub-unit because there is an interval between the formation of successive sub-units. The volume fraction of bainite depends on the totality of sheaves growing from different regions in the sample. Carbide precipitation events also influence the kinetics, primarily by removing carbon either from the residual austenite or from the supersaturated ferrite.

Little is known about the nucleation of bainite except that the activation energy for nucleation is directly proportional to the driving force for transformation. This is consistent with the theory for martensite nucleation. However, unlike martensite, carbon must partition into the austenite during bainite nucleation, although the nucleus then develops into a sub-unit which grows without diffusion.

The scale of individual plates of ferrite is too small to be resolved adequately using optical microscopy, which is capable only of revealing clusters of plates. Using higher resolution techniques such as photoemission electron microscopy it has been possible to study directly the progress of the bainite reaction. Not surprisingly, the lengthening of individual bainite platelets has been found to occur at a rate which is much faster than expected from a diffusion-controlled process. The growth rate is nevertheless much smaller than that of martensite, because the driving force for bainite formation is smaller due to the higher transformation temperatures involved. The platelets tend to grow at a constant rate but are usually stifled before they can traverse the austenite grain.

The lengthening rate of a sheaf is slower still, because of the delay caused by the need to repeatedly nucleate new sub-units. Nevertheless, sheaf lengthening rates are generally found to be about an order of magnitude higher than expected from carbon diffusion-controlled growth. Measurements have also been made of the thickening of bainite sheaves, a process which appears to be discontinuous, the thickness increasing in discrete steps of about 0.5 micrometer. These step heights correlate with the size of the sub-units observed using thin foil electron microscopy. The thickening process therefore depends on the rate at which sub-units are nucleated in adjacent locations within a sheaf.

These overall transformation characteristics, i.e. the change in the fraction of bainite with time, temperature, austenite grain structure and alloy chemistry are therefore best considered in terms of a TTT diagram. A simplified view is that the TTT diagram consists of two separable C-curves. The one at higher temperatures describes the evolution of diffusional transformation products such as ferrite and pearlite, whereas the lower C-curve represents displacive reactions such as Widmanstatten ferrite and bainite. In lean steels which transform rapidly, these two curves overlap so much that there is apparently just one curve which is the combination of all reactions. As the alloy concentration is increased to retard the decomposition of austenite, the two overlapping curves begin to become distinct, and a characteristic "bay" develops at about the B_S temperature in the TTT diagram. This bay is important in the design of some high-strength (ausformed) steels which have to be deformed in the austenitic condition at low temperatures before the onset of transformation.