喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有請放心下載，原稿可自行編輯修改=====================【QQ：1304139763可咨詢交流】喜歡就充值下載吧。。 喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有請放心下載，原稿可自行編輯修改=====================【QQ：1304139763可咨詢交流】喜歡就充值下載吧。。 喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有請放心下載，原稿可自行編輯修改=====================【QQ：1304139763可咨詢交流】喜歡就充值下載吧。。

Materials Science and Engineering A368 (2004) 2840 Texture evolution during equal channel angular extrusion Part I. Effect of route, number of passes and initial texture S. Ferrasse a, , V.M. Segal a , S.R. Kalidindi b , F. Alford a a Honeywell Electronic Materials, 15128 East Euclid Avenue, Spokane, WA 99216, USA b Department of Materials Engineering, Drexel University, 32nd Chesnut Street, Philadelphia, PA 19104, USA Received 4 July 2003; received in revised form 8 September 2003 Abstract It is shown that equal channel angular extrusion (ECAE) is an effective technique to control texture of metals and alloys. Two processing parameters, the route and number of passes, exert an important influence on texture evolution. Routes define orientations allowing the creation of numerous new components. Before four passes, depending on route and initial texture strength, all types of texture strength, from weak to very strong, are created, whereas after four passes, a global texture weakening is observed for all routes and medium to strong to very weak textures are produced. A simple Taylor model shows that crystallographic slip mechanically activated by simple shear is the governing mechanism for evolution of texture orientations. However, after four passes, the creation of submicron-grained structures with high misorientations is believed to limit crystallographic slip and weaken textures. 2003 Published by Elsevier B.V. Keywords: Equal channel angular extrusion; Texture of metals and alloys; Taylor model 1. Introduction Submicron-grained (SMG) materials (with a grain size less than 1H9262m) are attractive technical alternatives because they exhibit unusual physical and mechanical properties 1,2. Intensive plastic deformation has been proven as an effective method to produce SMG materials. Several tech- niques are capable of achieving the required strain levels 35. Of particular interest, the method of equal channel angular extrusion (ECAE) allows for the production of bulk pieces of SMG materials by simple shear 68. Mi- crostructures and properties of ECAE processed materials have been of high interest for the last 5 years 816. Very fine grain sizes (<1H9262m) result in unusual effects such as ideal plastic behavior, high strength, enhanced ductility and toughness, and low temperature superplasticity. Recently some studies have focused on textures of ECAE deformed metals 1737. Texture analysis is of great relevance to mechanisms of grain refinement during ECAE. At this stage a few particular textures corresponding to various cases of ECAE deformation and die angles have been investigated Corresponding author. or modeled. The intent of this paper is to present a sys- tematic study of texture formation in materials subjected to room temperature ECAE processing with a die angle of 90 and various numbers of passes and routes. This study clarifies the potential of ECAE technology to control texture. 2. Experimental 2.1. Original material and ECAE processing High purity Al0.5Cu alloy was initially cast, homogenized and solutionized. This single phase, high stacking fault en- ergy (SFE) alloy is a convenient material to analyze texture evolution under processing. All samples were submitted to two ECAE passes via route C 1012 followed by anneal- ing at 250 C, 1 h in order to produce an uniform and strong original texture with a recrystallized grain size of 60H9262m. Also, some samples were subsequently deformed by four passes route C and annealed at 300 C, 1 h to produce a weak texture with a recrystallized grain size of 100H9262m. Because of significant interest in texture development for flat prod- ucts, ECAE was performed for flat billets (Fig. 1) 38,39.A 0921-5093/$ see front matter 2003 Published by Elsevier B.V. doi:10.1016/j.msea.2003.09.077 S. Ferrasse et al. / Materials Science and Engineering A368 (2004) 2840 29 Y X Z BILLET AFTER ECAE ORIGINAL BILLET POSITION SHEAR PLANE Fig. 1. ECAE of flat billet. special note should be added on the definition of processing routes for this case. Different routes, which define selected shear planes and directions, may be attained by billet rota- tions about axis x, y or z (Fig. 1) at subsequent passes. Simi- larly to the elongated billets that are usually studied 40,in route A, no change in billet orientation occurs between each pass. In route B, the sample is alternatively rotated 90 about z axis at each pass (Fig. 1). In route D, the sample is rotated continuously +90 about the z axis at each pass. In route C, the sample is continuously rotated +180 about the z axis. 1 Note that z axis is the rotation axis of route B and D for flat billet instead of the x axis for elongated bil- let. For the strong initial texture, routes A, B and D were studied for number of passes 1, 2, 3, 4, 6 and 8 and, for route C, all passes from one through eight were observed. For the weak initial texture, only experiments via routes A and D from one to four passes were done. ECAE experi- ments were performed at room temperature and a low ram speed (0.75 mm/s). The processing of each route was care- fully controlled. Well lubricated 75 mm 75 mm 15 mm samples were processed in a die with sharp corners and an- gle of 90 (Fig. 1) that corresponds to a true strain per pass = 1.16. 2.2. Texture measurement Measurements of crystallographic texture were obtained with a Phillips XPert diffractometer. Textures were inves- tigated along the mid-thickness of a billet section parallel to the xy plane (Fig. 1). Partial pole figures on the 111, 200 and 220 planes were generated. The Beartex soft- ware 41 was used to process the raw data, correct each pole figure for background and defocusing, and calculate the orientation distribution function (ODF) 4143. An analy- 1 Some authors use the term B A for route B and B C for route D. In this case, routes B and D are not presented as basic routes but rather as dependent of routes A and C. sis of fibers (orientation tubes) was performed by examin- ing the ODF at constant values of 0, 5, 10, 15, 20, 22.5, 25, 30, 35, 40 and 45 according to the Roe/Matthies con- vention preferred in Beartex with Euler angles , , (see Table 1). Also, Beartex (option COMP) estimates the in- tegral intensity of the orientation distribution (OD) inside a given sphere in the orientation space that represents the volume fraction of grains possessing a given orientation in- side the chosen sphere. A 5 spread around the compo- nent was considered. Texture or OD index, which is the root-mean-squared (RMS) value of peak ODF, was also used to evaluate texture strength 44,45. It was assumed that OD index between 1 and 3 times random (t.r.) corresponds to random-to-weak textures, between 3 and 5 t.r. corresponds to weak-to-medium textures, between 5 and 20 t.r. corresponds to medium-to-strong textures and above 20 t.r. corresponds to extremely strong textures. Finally, for each pole figure, the extrusion axis of the last pass (x axis in Fig. 1) is always oriented along the x direction indicated in Fig. 4, and a per- pendicular to the plane of pole figure is parallel to z axis in Fig. 1. 2.3. Texture modeling A computer code based on a Taylor type crystal plasticity model was used to predict texture for simple shear defor- mation 45,46. It is applicable at low homologous tempera- tures where plastic deformation is mainly accomplished by crystallographic slip. Other deformation mechanisms such as shear banding, twinning, grain boundary sliding and rota- tion, as well as effects of grain shape and size, are not taken into consideration. Despite this simplicity, it was found that the model provides reasonable predictions for texture evolu- tion in single-phase, from medium to high stacking fault en- ergy cubic metals whatever the deformation mode 45,46. In simulations, 300 and 100 orientations were chosen to gen- erate the weak (near random) and strong original textures, respectively. 30 S. Ferrasse et al. / Materials Science and Engineering A368 (2004) 2840 Table 1 Basic description in the ODF space and inverse pole figure of principal fibers encountered for each route Major types of Fxy fiber Limits of location of fiber in (0 0 1) inverse pole figure (spreading due to ECAE route/pass). Name of fiber for each route (x = name of route) Range of Euler angles () (for typical ODF -cross sections) and corresponding range of plane Fx1 FA1, FB1, FD1 FBD1 (N = 2 for route B, D) FC1 (route C, N odd) (215255 1030 0) = (1 0 6)(1 0 2), (180230 1025 25) = (2 1 8)(2 1 4), (135190 520 45) = (1 1 8)(1 1 3) FC1 (route C, N even) (220255 3040 0) = (2 0 3)(1 0 1), (200250 2040 25) = (2 1 3)(2 1 2), (180190 1525 45) = (1 1 5)(1 1 2) Fx2 FA2, FB2, FC2, FD2 (2045 4050 22.5) = (2 1 2), (35015 3550 45) = (2 2 3)(3 3 4) Fx3 FA3, FB3, FC3, FD3, FBD3 (N = 2 for routes B, D) (95130 1535 0) = (1 0 3)(3 0 4), (90120 2535 15) = (7 2 10)(4 1 6), (80110 2545 25) = (2 1 3)(2 1 2), (35100 2035 45) = (1 1 3)(1 1 2) 3. Results 3.1. Initial textures The strong initial texture (OD index = 21.7 t.r.) contains principally three major orientations (Table 2 for N = 0). This atypical texture corresponds to the strongest textures encountered in traditional forming processes. The weak ini- tial texture has an OD index of 2.6 t.r. (Table 2 for N = 0). 3.2. Evolution of texture orientation Table 1 displays the definition of principal fibers found for each ECAE route. Tables 25 describe major orientations (with Euler angles , , (Roe/Matthies) and corresponding xyzuvw ideal description) and associated fibers. Texture orientation evolves similarly for both types of initial textures with some notable features. First, starting texture exerts a limited influence by shift- ing or weakening orientations for one and two passes (Tables 25; see route A and D in Fig. 2a and b). At a higher number of passes, influence of the original texture on orien- tation disappears. For example, for three and four passes via route D (Fig. 2c and d) inverse pole figures exhibit similar orientations either for a strong or random initial texture. Second, ECAE route has a tremendous impact on texture orientation especially at low number of passes. In fact, all orientations created with the four considered routes cover a significant area of the standard triangle (Tables 25, Figs. 25). For example, with a strong initial texture, as many orientations as (1 0 2)241 (route A), (2 1 2)425 (route D), (4 1 6)413 (route C), (3 0 4)463 (route C) and, (7 2 10)491 and (1 0 4)451 (route B) are created. Such orientations are fairly unusual compared to those obtained by traditional forming operations. Third, ODF analysis indicates that a limited group of fibers containing one or several major orientations are asso- ciated with each route irrespective of number of passes and original texture (Tables 15; Fig. 6). The number of fibers varies from three to six but only one or two fibers are pre- dominant for each route. These main fibers are noted FA3 and FA2 for route A (Fig. 6), FBD2 for routes B and D at two passes, FD1 for route D after three passes, FB3 and FB1 for route B after three passes, and FC3 and FC1 for route C (Table 1). The large number of fibers reflects the asymmetry of textures created during ECAE as evidenced by the (1 1 1) pole figure obtained after one pass for a random initial tex- ture (Fig. 5b). Its patterns are similar to a plane rolling tex- ture 19,45 with a rotation of 1015 about y axis, which is directly responsible for the observed asymmetry. At a higher number of passes, textures remain planar and asymmetric and gradually different from the rolling case. Fourth, there is also some similarity between the tex- tures created by different routes as evidenced by pole fig- ures (Figs. 5 and 6) and deformation fibers (Table 1). For a given fiber and cross-section in the ODF space, the effect of route on orientation is to change the angle and, in a lesser extent, the angle . As a result, since angles and S. Ferrasse et al. / Materials Science and Engineering A368 (2004) 2840 31 Table 2 Major orientations and corresponding fibers for route A Number passes N Major orientations (Notation: Euler angles ():xyzuvw: % total volume with 5 spread) Corresponding fiber Route A (strong initial texture) Original (N = 0) (10.9 54.7 45): (111)123:16 (105 26.5 0): (102)281:14 (110 24 26.5): (215)551: 9.3 N = 1 (119 26.5 0): (102)241: 17.62 FA3 (346 43.3 45): (223)212: 7.62 FA2 N = 2 (138 26.5 0): (102)221: 8.66 FA3 (31 36.7 26.5): (213)364: 8.6 FA2 N = 3 (126.7 26.5 0): (102)231: 7.45 FA3 (21 36.7 26.5): (213)243: 6.1 FA2 N = 4 (26.5 36.7 26.5): (213)264: 9.42 FA2 (138 26.5 0): (102)221: 4.62 FA3 (169 15.8 45): (115)321: 4.32 FA1 N = 6 (126.7 26.5 0): (102)231: 6.66 FA3 (228 33.7 0): (203)342: 5.8 FA1 (31 36.7 26.5): (213)364: 3.42 FA2 N = 8 All major orientations < 3.1 Route A (weak initial texture) N = 0 (80 25.2 45): (113)811 1: 4.3 N = 1 (0 46.7 45): (334)223: 5.8 FA2 (222 26.5 0): (102)221:5 FA1 (128 18.4 0): (103)341: 4.01 FA3 N = 2 (126.7 26.5 0): (102)231: 6.22 FA3 (26.5 48.2 26.5): (212)122: 5.4 FA2 (162 13.2 45): (116)421: 5.4 FA1 N = 4 (233 26.5 0): (102)231: 4.63 FA1 (136 19.5 45): (114)401: 4.54 FA1 (26.5 36.7 26.5): (213)364: 3.7 FA2 affect (xyz) plane and uvw directions, respectively, either the major orientation inside a fiber changes or a new fiber with new orientation is produced. The specific denomination mentioned earlier takes into account these similarities. A fiber name has the form Fxy, where a letter x corresponds to the considered route (respectively A, B, C, D) and a number y corresponds to a specific set of angles , . Three major types of fibers Fx1, Fx2 and Fx3 were identified as shown in Table 1. Each of these fibers exhibits similar values for the Table 3 Major orientations and corresponding fibers for route B Number of passes N Major orientations (Notation: Euler angles ():xyzuvw: % total volume with 5 spread) Corresponding fiber Route B (strong initial texture) N = 2 (0 48 2 6.5): (212)425: 24.24 FBD2 (216 15.8 45): (115)241: 8.07 FBD1 (138 26.5 0): (102)221: 5.04 FBD3 N = 3 (260 36 74): (2 7 10)941: 15.49 FB3 (118 18.4 90): (0 1 3)631: 5.23 FB1 N = 4 (100 36 16): (7 2 10)491: 12 FB3 (230.5 14 0): (104)451: 8.05 FB1 N = 6 (230.5 14 0): (104)451: 12.46 FB1 (100 36 16): (7 2 10)491: 10.2 FB3 N = 8 (230.5 14 0): (104)451: 9.19 FB1 (180 13.2 45): (116)331: 8.21 FB1 (100 36 16): (7 2 10)491: 7.48 FB3 set of angles , for most of the routes and passes (for ex- ample, from around (230, 0) to (180, 45) for Fx1(Table 1)), but specific values of angles and, to some extent, depend on deformation path. Table 1 shows the area of the inverse pole figure as well as the range of planes and Euler angles covered by each of these major fibers (for example, Fx1, when regrouping all studied routes corresponding to FA1, FB1, FC1, FD1 and FBD1 fibers). Fig. 6 shows fibers FA1, FA2 and FA3 in ODF space after the first pass. 32 S. Ferrasse et al. / Materials Science and Engineering A368 (2004) 2840 Table 4 Major orientations and corresponding fibers for route C Number of passes N Major orientations (Notation: Euler angles ():xyzuvw: % total volume with 5 spread) Corresponding fiber Route C (strong initial texture) N = 2 (0 34.5 14): (416)413: 43.3 FC3 (221.8 265 0): (102)221: 10.5 FC1 N = 3 (254 18.4 0): (103)3111: 7.5 FC1 (111.5 46.5 18.4): (313)231: 6.6 FC3 N = 4 (130 36.9 10): (304)463: 15.05 FC3 FC1 < 3.75 FC1 N = 5 (270 14 0): (104)010: 4.66 FC1 (26.5 48 26.5): (212)122: 2.54 FC3 N = 6 (110 36 16): (7 2 10)582: 11.6 FC3 (234 33.7 0): (203)352: 5.7 FC1 N = 7 (242 18.4 0): (103)361: 4.66 FC1 (188 11.4 45): (117)341: 3.36 FC3 N = 8 (136.5 18.4 0): (103)331: 14.71 FC3 (257 45 0): (101)8498: 8.75 FC1 Fifth, for any route, less variation in orientation is seen as the number of passes N augments. Some principal fibers and orientations are gradually selected and conserved as N increases. Texture evolution can then be described by sim- ple rules, which show how processing macromechanics in- fluences texture: For route A, material distorsion is close to monotonic, which results in a few gradual changes in orientations (Figs. 2 and 5, Table 2). Mainly, the same planes, (1 0 2) and (2 1 3), are involved. This type of linear evolution is similar to that found in rolling; For route D, after three passes, no significant changes in the ODF or pole figure is observed whatever the original texture (Fig. 2; Table 5). This status quo can be surprising since a rotation of 90 is continuously done at each pass. Table 5 Major orientations and corresponding fibers for route D Number of passes N Major orientations (Notation: Euler angles ():xyzuvw: % total volume with 5 spread) Corresponding fiber Route D (strong initial texture) N = 2 (0 48 26.5): (212)425: 24.24 FBD2 (216 15.8 45): (115)241: 8.07 FBD1 (138 26.5 0): (102)221: 5.04 FBD3 N = 3 (197 20.4 26.5): (216)221: 9.57 FD1 Major orientation in FD2, FD6, FD3 < 3 N = 4 (222 26.5 0): (102)221: 13.34 FD1 FD2, FD3, FD6 < 3.8 N = 6 (223.5 18.5 0): (103)331: 7.4 FD1 Major orientation in FD3, FD2, FD6 < 2.5 N = 8 Major orientation in FD1, FD2, FD3 3. A similar observation was made in 21. It can be argued that the amount of defor- mation involved at each pass is so high ( = 1.16) that any original influence is diminished. Even in route C, despite the inversion of simple shear and because of hysteresis, the 38 S. Ferrasse et al. / Materials Science and Engineering A368 (2004) 2840 Fig. 8. Inverse (0 0 1) pole figures predicted by modeling for a weak initial texture after two passes via route D (a), four passes via route D (b), and four passes via route A (c). components obtained after two passes differ from the origi- nal texture 22. Finally, as the number of passes increases, especially after four passes, less change in texture orientation occurs. There is rather a global spreading around some final orientations for each route that results from texture weak- ening and structural transformations as explained below. 4.2. Evolution of texture strength A global weakening of texture from medium to very weak is observed as the number of passes augments, especially af- ter three or four passes, whatever the initial texture. Textural randomization is enhanced by weaker initial textures 22, and, in decreasing order of influence, for routes D, A and C (in the last case for odd numbered passes). It is still possible to obtain medium-strong textures after four passes as shown in routes B or C (for even numbered passes) with a strong initial texture. Texture softening at increased passes was also found in 18,28,31,35. Contrary to experiments, com- puter simulation predicts continuous texture strengthening for all routes when number of passes augments. Predictions in orientation also become less accurate at high strains. These contradictory results prove that crystallographic slip is not the only mechanism responsible for texture evolution during ECAE 47,48. In particular, the simple Taylor code does not consider all microstructural changes occurring dur- ing ECAE 28,4548. The abundant literature concerning ECAE 837,47,48 shows that at high strains, a sub-micron structure is created (see part II, Fig. 1). Break up and refine- ment of microstructures is induced mechanically by dynamic recrystallization, which is believed to involve mechanisms such as grain/subgrain rotation, flow localization in shear bands and their intersections 9,10,12,15,33,4751. These mechanisms contribute to diminish texture strength after three passes. Basically, the areas belonging to each fiber at first passes are divided into highly stressed and misoriented SMG structures, which disperse more X-rays and weaken fibers. The occurrence of texture spreading at high deformations was reported for different processes including, for example, rolling after 9095% reduction 44,5255. Most of these observations were made for strains, which were not as large as in this study. As ECAE