A method for the calculation of retained austenite evolution during heat-treatment of low-alloy TRIP-assisted steels

Despite the critical effect of heat-treatment, and in particular of the isothermal bainitic treatment stage, on the amount and stability of retained austenite in the microstructure of low-alloy TRIP-assisted steels, determination of optimum heat-treatment conditions is still largely empirical and experiment-dependent. This work proposes a method by which it is possible to calculate the vol. fraction of retained austenite in the microstructure as a function of intercritical annealing temperature and isothermal bainitic treatment temperature and holding time. The method assumes diffusionless lengthening of bainitic ferrite (alpha(B)) plates in austenite (gamma), and subsequent thickness-wise C rejection from the alpha(B) plates to the adjacent gamma layers. The relative thickness of alpha(B) plates and adjacent gamma layers is determined by the T-o line of the transforming system at any given bainitic transformation temperature. The C-concentration profiles in gamma are calculated with respect to a local time-scale, referring to any random section of any random alpha(B) plate. Determination of the variation of C-concentration profiles with local time in gamma, together with the use of a simple austenite-retention criterion, allows the calculation of vol. fraction retained austenite (gamma(R)) as a function of transformation temperature and local time. Transition from local (calculation) time to actual heat treatment time is performed by introducing a time-scale factor, which depends on transformation temperature and initial C-content of alpha(B). The calculated behaviour of vol. fraction gamma(R) vs. bainitic holding time conforms to the well established, experimentally observed one: vol. fraction gamma(R) initially increases with holding time, reaches a maximum and decreases at longer holding times. According to calculated results, the decrease is attributed to the gradual homogenization of C inside the gamma layers, which leads to inadequate stabilization and transformation of gamma to martensite on quenching. As regards quantitative comparison with available experimental data, calculations show reasonable agreement. Certain refinements of the method, which are underway, are reported, in order to further improve quantitative results. Nevertheless, the method in its present form provides a calculational tool, by which the effect of different heat-treatment conditions or steel compositions can be examined and compared to each other, which can be a useful step towards optimizing alloy compositions and heat-treating processes.