Multislice image simulations of sheared needle-like precipitates in an Al-Mg-Si alloy.

The image contrast of sheared needle-like precipitates in the Al-Mg-Si alloy system is investigated with respect to shear-plane positions, the number of shear-planes, and the active matrix slip systems through multislice transmission electron microscopy image simulations and the frozen phonon approximation. It is found that annular dark field scanning transmission electron microscopy (ADF STEM) images are mostly affected by shear-planes within a distance ∼6-18 unit cells from the specimen surface, whereas about 5-10 equidistant shear-planes are required to produce clear differences in HRTEM images. The contrast of the images is affected by the Burgers vector of the slip, but not the slip plane. The simulation results are discussed and compared to experimental data. LAY DESCRIPTION: Pure aluminium is too soft to be viable in most structural applications, but this may be remedied by alloying the metal with various elements. Adding small amounts of silicon and magnesium to pure aluminium allows small particles to precipitate during heat treatment. These precipitates resist plastic deformation and can increase the strength of the alloy and make it viable for a range of industrial applications, such as automotive door panels and load-bearing profiles. However, if subjected to large loads, the precipitates are sheared and the strength of the alloy changes dynamically. Designing safe products such as cars or buildings require physically based predictions on this dynamical change. Developing models that can provide such predictions depend in turn on experimental observations of the shearing process. Because the precipitates are nm long, experimental observations must be done by transmission electron microscopy. However, understanding these results sometimes require computer simulations of atomic models. In this work, we have performed image simulations of various models of sheared precipitates and compared the results with earlier experiments. The simulations indicate that certain conditions must be met for the sheared precipitates to appear different from unsheared precipitates. These conditions are most likely to be met if precipitates are sheared several times in a relatively homogeneous manner. This is important for two reasons. First, a localized shearing process would lead to large dynamical changes in precipitate strength during deformation, and in turn drastically reduce the work hardening of the alloy. Secondly, a localized shearing process would have promoted earlier fracture and failure of the alloy during deformation. Finally, our results also show how different slip directions influences the images of precipitates. In the future, these influences can be used to further understand the shearing process of these precipitates. Hence, our results can be used to improve model predictions of strength, work hardening, and fracture. In turn, this may improve alloy design and reduce the use of prototype testing in, e.g. the automotive industry.