Evolution of the Microstructure of Al 6082 Alloy during Equal-Channel Angular Pressing

Transcription

1 Materials Science Forum Vols (5) pp online at 5 Trans Tech Publications, Switzerland Evolution of the Microstructure of Al 682 Alloy during Equal-Channel Angular Pressing J. Gubicza 1, Gy. Krállics 2, I. Schiller 1 and D. Malgin 3 1 Department of Solid State Physics, Eötvös University, Budapest, P.O.B. 32, H-1518, Hungary 2 Budapest University of Technology and Economics, Budapest, Hungary 3 DREHSDEN Ltd., Budapest, Hungary gubicza@ludens.elte.hu Keywords: Equal-Channel Angular Pressing, X-ray diffraction peak profile analysis, microstructure, dislocation density Abstract. A commercial Al-Mg-Si alloy (Al 682) was deformed by Equal-Channel Angular Pressing (ECAP) to produce bulk ultrafine-grained microstructure. The crystallite size distribution and the characteristic parameters of the dislocation structure were investigated by X-ray diffraction profile analysis. It was found that the crystallite size decreased and the dislocation density increased during ECAP deformation. The increase of the yield stress of the alloy was related to the increase of the dislocation density using the Taylor model. Introduction Severe plastic deformation (SPD) techniques are generally applied for obtaining ultrafine-grained (UFG) microstructure in bulk metals and alloys [1]. Equal-Channel Angular Pressing (ECAP) is the most often used method among SPD procedures since it results in homogeneous nanostructure without changing the dimensions of the bulk specimen [2,3]. The UFG materials produced by ECAP have an attractive combination of high strength and good ductility due to their low contamination and unique structures [4]. For understanding the mechanical behavior of materials produced by ECAP it is necessary to characterize their microstructure. From the point of view of industrial application, the investigation of ECAP-induced evolution of microstructure in commercial alloys is very important. X-ray diffraction profile analysis is an effective tool for studying the microstructure of UFG materials. The standard methods of profile analysis determine the apparent crystallite size and the mean square strain from the full widths at half maximum (FWHM), the integral breadths or the Fourier coefficients of the profiles [5,6]. In SPD processed materials where the lattice distortions are primarily caused by dislocations the mean square strain can be expressed in terms of the characteristic parameters of the dislocation structure [7,8]. In these formulas the anisotropic strain broadening of the peak profiles is taken into account by the contrast (or orientation) factors of dislocations [9-11]. In the last few years, a fast development in computing made it possible to work out procedures for determining the parameters of the microstructure by fitting the whole diffraction profiles [12 14]. In the recently elaborated Multiple Whole Profile (MWP) fitting method, the measured profiles or their Fourier transforms are fitted by theoretical functions calculated on the basis of a model of the microstructure [13,14]. This procedure enables to determine both the crystallite size distribution and the dislocation structure in UFG materials. In this study the evolution of UFG microstructure in a commercial Al-Mg-Si alloy (Al 682) during ECAP deformation is investigated. The crystallite size and the dislocation density are determined by X-ray diffraction profile analysis. The increase of the yield stress during ECAP deformation is related to the evolution of microstructure. Licensed to Gubicza (gubicza@ludens.elte.hu) - Eötvös University - Hungary All rights reserved. No part of the contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the publisher: Trans Tech Publications Ltd, Switzerland, (ID: /9/4,1:35:6)

2 454 Materials Science, Testing and Informatics IV Experimentals The material used in this study was a commercial Al-Mg-Si alloy (Al 682). The main components of the alloy are Al (97 %), Si ( %), Mg ( %) and Mn (.4-1 %). Before the ECAP deformation, the material was aged on ºС for 4 minutes. The annealed specimen was regarded as the as-received material. Cylindrical billets of 15 mm in diameter and 145 mm in length were pressed through the ECAP die with 9 intersecting channels. One, four and eight passes were performed by route A (the billet was not rotated around its longitudinal axis between intermediate passes) at room temperature at constant displacement rate of 8 mm/min. The applied load increased from 12 to 18 kn with increasing the number of ECAP passes because of the strain hardening of the alloy. The ratio of length to diameter of the specimens was relatively high (more than 9) compared to previous experiments of other authors where this ratio is usually between 2 and 5 [15]. The effect of the technological conditions (e.g. displacement rate, temperature, friction) on the mechanical properties of the ECA pressed alloy is studied in Ref. [16]. For investigation of the orientation of the crystallites in the specimens, X-ray diffraction patterns were recorded on both longitudinal and cross sections of the cylindrical billets. The X-ray diffractograms for the two sections were measured by a Philips X pert powder diffractometer using a Cu anode and a pyrolitic graphite secondary monochromator. For studying the microstructure of the as-received and ECA pressed materials, the peak profiles were also measured on the cross section by a high-resolution double-crystal diffractometer (Nonius, FR 591) using Cu Kα 1 radiation. The peak profiles were evaluated by the Multiple Whole Profile (MWP) fitting procedure described in details in Refs. [13,14]. In this method, the Fourier coefficients of the experimental profiles are fitted by the product of the theoretical Fourier transforms of size and strain peak profiles [13,14]. The theoretical functions used in the fitting are calculated on the basis of a model of the microstructure. In this model, the crystallites have spherical shape and log-normal size distribution, and the lattice strains are assumed to be caused by dislocations. The method gives the crystallite size distribution as well as the density and the arrangement of dislocations. In this study the volumeweighted mean crystallite size (<x> vol ), the dislocation density (ρ) and the dislocation arrangement parameter (M) are presented for the as-received and the ECA pressed specimens. Results and discussion Fig. 1 shows the X-ray diffraction patterns recorded on the longitudinal and cross sections of the asreceived specimen (a and c) and those obtained for the sample deformed by pass (b and d). The difference between the relative intensities of the Bragg peaks measured on the two sections of the as-received material reveals a strong texture (see Figs. 1a and c). This texture was developed when the commercial Al-Mg-Si alloy (Al 682) was casted and extruded by the producer. After 1 ECAP pass the difference between the X-ray diffraction patterns obtained on the two sections was diminished (see Figs. 1b and d) and after 8 passes almost completely disappeared. This indicates that during ECAP deformation the orientation distribution of the crystallites on the cross and longitudinal sections became similar. However, the orientation distributions of the crystallites on the different longitudinal sections can differ from each other and the correct investigation of the texture requires the record of X-ray diffraction pole figures. The Bragg peaks measured on the cross sections by high resolution diffractometer were evaluated by MWP fitting procedure. Fig. 2 shows the Fourier coefficients of the measured profiles (open circles) and the fitted theoretical Fourier transforms (solid line) for the specimen deformed by pass. The difference between the measured and fitted values is also plotted in the figure. The volume-weighted mean crystallite size (<x> vol ) and the dislocation density (ρ) are shown as a function of the number of ECAP passes in Fig. 3. Nanosized microstructure (<x> vol <1 nm) with high dislocation density was achieved even after 1 pass. The dislocation density increased with the increase of ECAP deformation up to 4 passes. The dimensionless dislocation arrangement parameter, M, has a value of 4.±.4 for the as-

3 Materials Science Forum Vols received specimen and it decreased to 2.2±.3 after 8 ECAP passes. This indicates that the dipole character of the dislocation structure became stronger with the increase of deformation. 2.5x1 6 2.x1 6 as-received cross section 16 cross section 1.5x1 6 1.x1 6 5.x a) b) 8.x1 4 6.x1 4 as-received longitudinal section 1 longitudinal section 4.x x c) d) Figure 1: The X-ray diffraction patterns for the longitudinal and cross sections of the as-received specimen (a and c) and for the sample deformed by pass (b and d). Tensile tests were carried out on the as-received and the ECA pressed specimens at room temperature. The yield stress as a function of ECAP passes are plotted in Fig. 4. The yield stress increased with the increase of ECAP deformation up to 4 passes. The yield stress obtained after 4 passes is about two times higher than that determined in the as-received state. The increment of the yield stress during deformation can be expressed by the increase of the dislocation density using the Taylor-model: σ Taylor =σ +αm T Gbρ 1/2, (1) where σ Taylor is the yield stress, σ is the friction stress (σ =25 MPa [17]), α is a constant (α=.33 is taken), G is the shear modulus (G=26 GPa [17]), b is the length of the Burgers vector of dislocations (b=.2865 nm) and M T is the Taylor factor (M T =3 for untextured polycrystalline materials [18]). The measured yield stress versus the values of σ Taylor calculated from the dislocation density according to Eq. 1 are plotted in Fig. 5. The error bar of σ Taylor for the as-received specimen is much higher than those for the ECA pressed samples because of the uncertainty of M T. The origin of this uncertainty is the texture developed in the as-received state of the material. Former studies [11,18] have shown

4 456 Materials Science, Testing and Informatics IV that the texture in Al alloys could reduce the Taylor factor even to 2.2 which can result in the decrease of the calculated value of the yield stress. Fig. 5 shows that the measured yield stress values are somewhat higher than those calculated according to Eq. 1. This deviation can be explained by the fact that the Taylor model takes into account only the dislocation-dislocation interaction in strengthening. At the same time the yield stress can also increase by solute hardening caused by alloying atoms (mainly Mg, Si and Mn) in Al matrix. Beside the Al solid solution a small amount of Mg 2 Si and Mn 12 Si 7 Al 5 phases were identified by X-ray diffraction. The strongest lines of these phases can be seen as the small peaks between and reflections of Al solid solution in Figs. 1b and d. The appearance of these phases results in precipitation hardening of the alloy which increases further the yield stress. Nevertheless, the above analysis shows that the high dislocation density is most likely a dominant factor in determining the strength of UFG Al 682 alloy in this study. Fourier transform 1..5 L=12 nm 422 Figure 2: The Fourier coefficients of the measured profiles (open circles) and the fitted theoretical Fourier transforms (solid line) for the specimen deformed by 1 ECAP pass. <x> vol [nm] Number of ECAP passes Figure 3: The volume-weighted mean crystallite size (<x> vol ) and the dislocation density (ρ) as a function of the number of ECAP passes. 6 3 ρ [1 14 m -2 ] 3 Yield stress [MPa] 1 Yield stress [MPa] ECAP 4 ECAP As-received Number of ECAP passes 1 3 σ Taylor [MPa] Figure 4: The yield stress as a function of ECAP passes. Figure 5: The measured yield stress vs. the values of σ Taylor calculated by the Taylor model.

5 Materials Science Forum Vols Summary Equal-Channel Angular Pressing (ECAP) was successfully applied to produce ultrafine-grained microstructure in a commercial Al-Mg-Si alloy (Al 682). The ratio of length to diameter of the specimen was extremely high (more than 9). The difference between the orientation distributions of the crystallites obtained on the longitudinal and cross sections of the as-received billet was diminished during ECAP. The crystallite size decreased and the dislocation density increased as a result of ECAP deformation. The increase of the yield stress of the alloy can be explained by the increase of the dislocation density using the Taylor-model. The deviation of the measured yield stress from the calculated stress values can be attributed to the solid solution and precipitation hardenings. Acknowledgements This work was supported by the Hungarian Scientific Research Fund, OTKA, Grant Nos. F-4757, T and T References [1] R.Z. Valiev, R.K. Islamgaliev and I.V. Alexandrov: Prog. Mat. Sci. Vol. 45 (), p. 13 [2] Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto and T.G. Langdon: Scripta Mater. Vol. 35 (1996), p. 143 [3] L.S. Toth: Adv. Eng. Mat. Vol. 5 (3), p. 38 [4] V.V. Stolyarov, Y.T. Zhu, I.V. Alexandrov, T.C. Lowe and R.Z. Valiev: Mater. Sci. Eng. A Vol. 33 (1), p. 82 [5] G.K. Williamson and W.H. Hall: Acta Metall. Vol. 1 (1953), p. 22 [6] B.E. Warren and B.L. Averbach: J. Appl. Phys. Vol. 21 (195), p. 595 [7] M.A. Krivoglaz: Theory of X-ray and Thermal Neutron Scattering by Real Crystals (Plenum Press, New York 1996). [8] M. Wilkens: Phys. Stat. Sol. (a) Vol. 2 (197), p. 359 [9] R. Kuzel jr. and P. Klimanek: J. Appl. Cryst. Vol. 21 (1988), p. 363 [1] T. Ungár and A. Borbély: Appl. Phys. Lett. Vol. 69 (1996), p [11] A. Borbély, J.H. Driver and T. Ungár: Acta Materialia Vol. 48, (), p. 5 [12] P. Scardi and M. Leoni: Acta Cryst. A Vol. 58 (2), p. 19 [13] T. Ungár, J. Gubicza, G. Ribárik and A. Borbély: J. Appl. Cryst. Vol. 34 (1), p. 298 [14] G. Ribárik, T. Ungár and J. Gubicza: J. Appl. Cryst. Vol. 34 (1), p. 669 [15] Z. Horita, T. Fujinami, M. Nemoto and T. G. Langdon: J. Mat. Proc. Techn. Vol. 117 (1), p. 288 [16] G. Krállics, D. Malgin, Z. Szeles and A. Fodor: Mater. Sci. Forum (in this issue) [17] X. Duan and T. Sheppard: J. Mat. Proc. Techn. Vol (2), p. 179 [18] B. Clausen, T. Lorentzeni and T. Leffers: Acta mater. Vol. 46 (1998), p. 387