Early Stages of Ageing in Al-Mg-Si Alloys

Abstract

Natural ageing is known to have a negative effect on the formability and bake-hardening response of Al-Mg-Si alloys. This is attributed to the formation of Mg and Si clusters during natural ageing. The clustering process was the subject of many studies in the literature, however, the formation mechanism and kinetics of it, continues to be poorly understood. The aim of this project is to shed some light on the cluster formation mechanism and measure clustering kinetics at low temperatures. A series of electrical resistivity measurements, positron annihilation lifetime spectroscopy and hardness tests were performed on samples aged over the temperature range of -20 to 50°C following solution treatment at temperatures of 525 and 560°C. A very good correlation between the results of various techniques was observed. In addition, three different stages in the clustering process were detected. Not surprisingly it was found that the excess quenched-in vacancies are the key players in the cluster formation process. In the first stage, annihilation of near-sink vacancies occurs while other vacancies start to bind with solute atoms and form clusters. In the second stage, clustering continues to take place but its rate slows down since the effective diffusion coefficient of vacancies decreases as they bind with more solute. Finally, the clustering process enters the third stage with much slower kinetics. Interestingly, positron annihilation lifetime also reaches a constant value at the beginning of stage III which suggests the stabilization of vacancies. Two hypothesis were then developed to explain the existence of stage III: one based on the immobilization of vacancies due to the increased binding with solute atoms and another one which considers the overlapping of solute diffusion profiles around the clusters. Finally it was shown that the resistivity change in stage II can be used to find the activation energy of clustering which is calculated to be approximately 46 kJ/mol. This is very close to the migration energy of vacancies and Mg atoms. Thus it was concluded that migration of these species is the major rate controlling parameter for the clustering process.