Modeling of the mechanical behavior of low carbon martensite based on the stochastic approach of the microstructure heterogeneity

Recent studies have suggested that the strain hardening of low carbon martensitic steels can be better understood by considering the important microstructural heterogeneities, arising from the progressive nature of the martensitic transformation. In this work, these heterogeneities are systematically analyzed in three martensitic steels with different carbon contents, focusing on the correlated distributions of dislocation density and martensitic domain size. From these characterizations, the local flow strength distributions were obtained at the scale of the microstructure. A Masing-like micromechanical model was then used to predict the tensile behavior by combining the calculated yield strength distribution with a proportional distribution of deviatoric stresses. The model successfully reproduces the tensile behavior of the studied steels in the as-quenched state and, for the first time, quantitatively accounts for the role of microstructural heterogeneities. Moreover, the subsequent compressive test is successfully modeled without the introduction of any supplementary parameter, highlighting that the model effectively describes the essentially kinematic nature of strain-hardening in martensitic steels.