Impact of die design and bearing geometry on grain size and PCG formation during extrusion of AA6082 aluminum alloy
Marco Negozio, Sara Di Donato, Riccardo Pelaccia, Adrian H. A. Lutey, Daniele Carosi, Barbara Reggiani, Alessandro Morri, Lorenzo Donati
Abstract
• Examines the role of die geometry in PCG formation during AA6082 extrusion. • Evaluates PCG thickness impact on mechanical properties. • Combines FEM simulations and experiments to analyze microstructural evolution. • Validates predictive FEM model for grain size and PCG thickness. • Provides industrial tools to optimize extrusion process and reduce PCG defects. Grain size and formation of the Peripheral Coarse Grain (PCG) defect influence the mechanical and crash properties of extruded profiles. Controlling microstructural evolution during the extrusion of 6XXX series aluminum alloys is therefore essential to ensure the performance of structural components. In this work, three profiles with the same nominal geometry were extruded with a die comprising three different bearing geometries to create different extrusion conditions. Each profile was analyzed experimentally to gather data on the microstructure and mechanical properties. Bulge testing revealed that Profile 2, with the thickest PCG layer (490–1150 µm), exhibited worse mechanical performance, with a hoop strain at fracture of 0.08 and a peak load of 51.5 kN, compared to Profiles 1 and 3, which had higher hoop strains (0.13 and 0.14) and peak loads (56.1 and 57.6 kN, respectively). Finite Element Method (FEM) simulations of the extrusion process were carried out using Qform Extrusion UK with a post-processing subroutine developed and implemented to calculate additional parameters such as the stored energy, percentage dynamic recrystallization, grain size, and PCG formation based on standard output parameters from the simulation including strain, temperature and strain rate. The simulation demonstrated that the highest strain rate (40–220 s ‒1 ) and stored energy (150,000–440,000 J m ‒3 ) in Profile 2 led to the thickest PCG layer. Based on these results, the proposed predictive model was validated against experimental data, demonstrating high accuracy in predicting PCG thickness and grain size while effectively capturing the influence of process parameters on microstructural evolution.