A novel method of experimental determination of grain stresses and critical resolved shear stresses for slip and twin systems in a magnesium alloy

TL;DR

This paper introduces a new neutron diffraction method for directly measuring grain stresses and critical resolved shear stresses in magnesium alloys, enabling model-independent analysis and uncertainty quantification.

Contribution

The novel method allows direct, model-free determination of grain stresses and CRSSs, including uncertainty estimation, improving understanding of anisotropic behavior in magnesium alloys.

Findings

Direct measurement of grain stresses and CRSSs in magnesium alloy AZ31.

Quantification of uncertainties in CRSS and local stresses.

Validation of elastic-plastic self-consistent model parameters.

Abstract

A novel original method of determination of stresses and critical resolved shear stresses (CRSSs) using neutron diffraction was proposed. In this method, based on the crystallite group method, the lattice strains were measured in different directions and using different reflections hkl during uniaxial deformation of magnesium alloy AZ31. The advantage of this method is that the stresses for groups of grains having similar orientations can be determined directly from measurement without any models used for data interpretation. The obtained results are unambiguous and do not depend on the models assumptions as in previous works. Moreover, it was possible for the first time to determine the uncertainty of the measured CRSS values and local stresses at groups of grains. The used methodology allowed for the determination of stress partitioning between grains having different orientations and for an explanation of the anisotropic mechanical behaviour of the strongly textured alloy. Finally, the CRSS values allowed for the validation of the type of intergranular interaction assumed in the elastic-plastic self-consistent model and for a significant reduction of the number of unknown parameters when the model is adjusted to the experimental data.