Examination of carbon partitioning into austenite during tempering of bainite

The redistribution of carbon after tempering of a novel nanocrystalline bainitic steel consisting of a mixture of supersaturated ferrite and retained austenite has been analyzed by atom probe tomography. No direct evidence supporting the additional carbon enrichment of austenite beyond that initially achieved during the bainite heat treatment was obtained during subsequent tempering of this high carbon, high silicon steel. The martensite transformation is universally accepted to involve diffusionless growth of ferrite, which is supersaturated with carbon at low temperature. For many years, it has been recognized that carbon may partition from martensite to austenite after transformation, because the solubility of carbon is much greater in austenite [1,2]. Although the existence of carbon-enriched retained austenite in martensitic steel microstructures has been known for some time [3], the process of partitioning from the supersaturated martensite to the untransformed austenite has received little attention in quenched steels. Recently, a new processing concept known as quenching and partitioning (Q&P) has been proposed [4-6] for designing high strength, ductile steels [7]. This process involves quenching austenite below the martensite-start temperature, followed by a partitioning treatment to enrich the remaining austenite with carbon, thereby stabilizing it to room temperature [4-7]. Clarke et al. [8] proposed two possible mechanisms for austenite carbon enrichment of a Fe0.19 wt.% C-1.59% Mn-1.63% Si steel subjected to Q&P processing, including: (i) the partitioning of carbon to untransformed austenite from carbon supersaturated martensite and (ii) the carbon enrichment of austenite associated with the formation of carbide-free bainite. Theoretical calculations [8] showed the partitioning of carbon to austenite from the initial martensite was more consistent with the experimental austenite fraction results. Deviation from the predicted results suggested the loss of carbon atoms to competing processes, such as carbide precipitation or carbon segregation to dislocations [7]. The details of carbon partitioning during the bainite transformation have traditionally been somewhat more controversial. However, the determination of the carbon concentrations in bainitic ferrite by atom probe tomography (APT) in a high carbon,