Influência da laminação assimétrica nas propriedades mecânicas do alumínio AA 1050
Abstract
Aluminum sheets are produced by rolling followed by annealing. However when submitted to deep drawing earing appears, caused by the plastic anisotropy resulting from the production process. After annealing the main texture is cube {001} , that is recognized as being the cause of this heterogeneity during deep drawing. The literature shows that when shear stress is applied in the deformation process, it leads to texture changes. In this study asymmetric rolling (AR) was used as a technique to produce shear. The shear stress is introduced by the different velocities between the upper and bottom rolls and in this study this was achieved by using roll radius relations (r1/r2) of 1,5 and 2. Rolling reductions of 50% in thickness were applied to aluminum AA1050 sheets. The conventional rolling (CR) was compared to the asymmetric rolling (AR), at two different reduction rates: 5% and 10%. The crystallographic textures were obtained by xray diffraction. Finite element analysis, using the DEFORM software, was used to analyze the effective strain distribution throughout the thickness as well as its components: normal strain, shear strain and rigid body rotation. The samples were annealed in a furnace with 350°C for 05, 10, 15, 20 and 60 minutes. The microstructure was characterized by optical microscopy, electron back scatter diffraction and x-ray diffraction. The plastic anisotropy (Lankford Parameter) was measured by tensile experiments at three different sheet directions and by the Erichsen test. The deformed samples’ microstructure was analyzed at the surface near to the upper roll and at half of the thickness. For the CR the main components were brass (Bs) {011} , Goss (Gs) {011} and copper (Cu) {112} , with 8.8 intensity at the central layer, and 4.5 at the surface. For AR samples the was more random at the surface of the samples with 5% of reduction per pass, added to a component of rotation in the normal direction, what resulted in cube and rotated cube textures or near to these orientations, generating a type of fiber {100}//ND. The maximum intensities for the (r1/r2) of 1,5 and 2 were 3 and 4, respectively. For the samples with 10% of reduction per pass the rolled texture was still presented, with a more intense rotation in the transversal direction related to the rolling direction and shear texture components {100}//ND and . The maximum intensities were 3 and 3.5 , for the (r1/r2) of 1,5 and 2, respectively. In the center layer of the samples with 5% of reduction per pass for (r1/r2) of 1,5 and 2 showed a intensity of 5.26 and 6.56, respectively and the strongest shear texture was rotated Goss (C). (011)[0-1- 1] The samples with 10% reduction per pass showed the greatest reduction of intensity with 3.05 and 3.63, for the (r1/r2) of 1,5 and 2 respectively, and the highest intensity was related to rotate the Goss (011)[0-1-1] component. In the pole figures rotations around the transversal direction (TD) and the normal direction (ND) were observed. Using the finite element analysis the rotation around the TD and ND were quantified and its variation across the thickness were analyzed. The rigid body rotation is superposed to shear , which leads to the observed texture gradients. The rotation around TD is imposed by the velocity difference between top and bottom roll, whereas the ND rotation is imposed by the experimental configuration, which permit variation of the sample alignment at the roll mill entrance. This was stronger for the 5% reduction rate and more concentrated at the samples surface. After 05 minutes the annealed samples were already recrystallized , after 60 minutes the grain average size was 30µm, and hardness 21HV. The annealed texture for the CR sample showed the greatest concentration off Cube texture {001} and intensity of 8.08 times the random. For the AR samples with 5% reduction per pass the intensities for the (r1/r2) of 1,5 and 2 was 5.88 and 6.56, respectively, and for the 10% reduction per pass 2.96 and 2.85, respectively. The AR decreases the annealed texture. In the samples of 5% reduction per pass the most intense shear texture was rotated Goss, the 10% reduction per pass did not have a predominant component. The Lankford parameters showed less anisotropy for the annealed samples with 10% reduction per pass. Based on the values of anisotropy and hardening exponent for each sample, the Limiting Rate of Drawing was calculated. The AR got a superior values than the CR ones, indicating an improvement of the drawability.