Rapid assessment of the recyclability of Al alloys as a function of Fe content

Aluminum is an attractive sustainable material due to its use in several key sectors and can be continuously recycled while retaining its properties, making it an excellent candidate to meet the carbon neutrality and circular economy requirements of the European Green Deal. Recycling is often associated with introduction of impurities, such as increased Fe content affecting material performance. To decrease time, cost and energy, it is imperative to investigate the effect of increasing Fe content in Al-Mg-Si alloys and ways to manage the increase without the need to remove the impurities.

This study focuses on extruded AA6082 Al alloys (Al-0.7Mg-1.0Si-0.5Mn-xFe wt.%). The influence of variations in Fe content on alloy properties was investigated by simulation and experiment. As a first step, high-throughput computational screening was performed on hundreds of chemical compositions generated as a function of Fe content. The CALPHAD method was used for equilibrium and non-equilibrium simulations to evaluate key parameters including phases, solidification intervals, thermophysical parameters and mechanical properties such as yield strength. Subsequent sensitivity and uncertainty analysis allowed the effects of varying Fe content on microstructural, thermo-physical and mechanical properties to be addressed, as well as identifying Fe tolerance ranges for alloy properties with low sensitivity to Fe variations.

The second step involved experimental investigation on two AA6082 alloys with Fe content of 0.18 and 0.35 wt.%. The mechanical properties, especially fracture response, were studied by uniaxial tensile testing and VDA testing. SEM and micro-CT were conducted to characterize the size, number density and spatial distribution of the constituent particles, which potentially affect fracture behavior. Quantitative analysis for the constituent particles was accelerated by the automation for segmentation using machine learning. The fracture response was then related to the degree of clustering of the constituent particles.