Predicting elastic anisotropy of dual-phase steels based on crystal mechanics and microstructure

This work utilizes mean-field self-consistent and full-field fast Fourier transform-based homogenizations to study the effective elastic behavior of several steels: three dual-phase (DP), DP 590, DP 980, and DP 1180, and one martensitic (MS), MS 1700. Crystallographic textures and phase fractions of these steels are characterized using electron microscopy along with electron-backscattered diffraction to initialize the models. A comprehensive set of Young's modulus and Poisson's ratio data, measured at the ambient temperature as a function of orientation with respect to the rolling direction for each steel sheet, is used to calibrate and validate the models. The calibration of the models enabled us to estimate the single crystal elastic constants for both the martensitic phase and ferrite, while calculating the orientation dependent effective properties. Half of the data was used in the calibration. Subsequent predictions of the orientation dependent effective elastic properties for the remaining data verified that the estimated single crystal properties are reliable. As the steels exhibit a different level of anisotropy in their effective behavior, good predictions allowed us to discuss the role of texture, grain structure, phase fraction and distribution on the effective properties. The results of this work represent a significant incentive to introduce elastic anisotropy in numerical tools for simulating metal forming processes of dual-phase steels, in particular those processes involving springback, using the texture informed crystal mechanics-based models to more accurately estimate the effective elastic properties required by such simulations.