1
大量變形過程晶粒內高角邊界之產生
The For mation of Intr agr anular High Angle Boundar ies Dur ing Lar ge
Str ain Defor mation
計畫編號：NSC89-2216-E-110-038 執行期限：89 年 8 月 1 日至 90 年 7 月 31 日 主持人：高伯威 中山大學材料科學研究所 計畫參與人員：孫佩鈴、王允俊 中山大學材料科學研 中文摘要 本研究利用等徑轉角擠型以 90o和 120o兩種模具擠製 1050 商用純鋁。我們利 用穿透式電子顯微鏡(TEM)及掃描式電子 顯微鏡(SEM)之電子背向繞射(EBSD)分析微 組織之演化，並對兩種模具產生的組織作詳細 比較。 關鍵詞： 高角邊界，大量塑性變形，電子背 向繞射 Abstract
Compercial purity aluminum (1050 Al) was subjected equal channel angular extrusion (ECAE) by route C with the intersection of channel angles of 90o and 120o. TEM and EBSD (electron back scattered diffraction) were used to examine the evolution of these structures. The effect of die angle on the microstructure evolution was compared.
Keywords：high angle boundaries, large strain
deformation, electron
backscattered diffraction(EBSD)
1. Introduction
Severe plastic deformation (SPD) methods have been shown to be very effective in refining the grain size to the submicrometer scale. Valiev et al. [1] recently have a very good review on the processing techniques, microstructures and properties of the submicron grained materials resulted from severe plastic deformation. One of the widely used
techniques is equal channel angular extrusion (ECAE), which involves subjecting a material to severe plastic deformation repeatedly by shearing a billet in a die with two channels of identical cross-section that meet at an angle [2]. Two important parameters of the process are the die angle, which determines the shear strain per pass through the die [3], and the processing route, which determines the strain path [4, 5].
The ECAE mold contains two channels with equal cross section meet at an arbitrary angle Φ and an angle Ψ delineates outer curvature of the two channels. The strain introduced in each pass is determined by these two angles [6]. However, the die angle determines not only the strain applied per pass but also the orientation of the shear plane. It has been suggested that the evolution of microstructure in ECAE with different processing routes and die angles may be attributed to the changing of the orientation of the shear plane associated with them [4, 7]. Only with route C, the same shear plane is maintained in each pass [7]. It is the objective of this work to study the effect of strain per pass on the microstructure development in ECAE. By the use of route C and different die angles, the effect of strain per pass can be investigated without the interference of
shear plane intersection. The
microstructure was characterized by using both transmission electron microscopy (TEM) and electron back scattered diffraction (EBSD) in scanning electron microscopy (SEM).
2
2. Exper imental
Commercial pure aluminum AA1050 (99.5%), which was homogenized at 893 K for 12 hours then cooled slowly, was used in this study. The initial grain size is about 330 μm. The different dies (Φ=90o
, Ψ=20o and Φ =120o, Ψ =0o) were used in this study. The von Mises equivalent strains are 1.05 and 0.67 for the 90o and 120o die, respectively [6]. The ECAE process was carried out at room temperature with MoS2
as lubricant. The material was deformed to total strain of ~8.
The coordinate used to define the orientation of the specimen deformed by ECAE is shown in Fig.1. It has been recognized from previous studies that the evolution of microstructure is more rapid on X-plane than on Y-plane. At lower strains (<4), only the X-plane was examined. Specimens were also examined by SEM (JEOL 6400) equipped with electron back scattered pattern (EBSP) analysis system
(Opal, Oxford Instrument). The
measurements were made along lines parallel to the z-axis with a step size about 500 nm. For each specimen, measurements were made on five well separated lines, that each line covers 120 µm length. The specimens for EBSP analysis were annealed at 373 K for 1800 s to improve the quality of the Kikuchi patterns. Finally, the EBSP specimens were electropolished in STRUERS A2 electrolyte for about 30 s.
At higher strains (4 and 8), the specimens were examined by TEM (Philips CM 200). The microstructure on of both the X- and the Y-plane were examined and the misorientation angles of boundaries observed on the Y-plane were also determined. The misorientation angles across boundaries were measured by analyzing Kikuchi pattern to obtain the crystal orientations on the two sides of boundaries [8]. For the determination of misorientation angle distribution, more than six areas of each condition were examined. About 150 grains were measured in each specimen and over 250 boundaries were analyzed.
3. Results and Discussion
3.1 High angle boundaries developed at strain < 4
The local variation of crystallographic orientation was determined by the EBSP analysis. After one ECAE pass with both die angles, large variations in crystallographic orientation were observed on a scale, which is small compared to the initial grain size. It was noticed that for the same specimen, the misorientation varies significantly on different line scanned. It has been pointed that the microstructure developed by heavy deformation depends strongly on the grain orientation [9]. As shown before, the strain imposed per pass by the 90o die is higher than that by the 120o die. The observations showed that on the average, the specimen deformed one pass by the 90o die has higher frequency of high angle (>30o) misorientation. The EBSD study also showed that the 90o die results in higher rate of evolution of high angle misorientation at lower strain range (<4).
With ECAE route C, the shear strain is reversed every alternate pass. The original grain shape can be roughly reconstructed after even number of passes. Thus the formation of high angle boundaries can be mainly attributed to the process of grain subdivision [9].
3.2 Microstructure developed at strain of ~4
Because of the anisotropic
characteristics of the microstructure developed in ECAE, both the X-plane and the Y-plane were examined. Quantitative characterization was performed on the boundaries observed on the Y-plane by the use of TEM Kikuchi pattern analysis [8]. The deformed structures produced by both die angles will be compared at two strain levels, ~4 and ~8. At these strains, the materials processed by both dies experience fully redundant deformation such that they can be compared on the same base.
3
The microstructure on the X-plane of specimens produced by 90o die is quite heterogeneous at strain of 4. The main feature is the banded structure, but there are areas, where the structure is dominated by nearly equiaxed subgrains (grains). The banded structure is nearly parallel to the Y-axis. In the banded structure, the dislocation densities are still very high in some areas and cell structures appear frequently. For specimens deformed by the 120o die, the dominant feature on the X-plane is also the banded structure.
On the Y-plane of specimens produced by both dies, the dominant feature is a mixture of banded structure and equiaxed subgrains (grains). Generally, the microstructure on the Y-plane is quite similar to that on the X-plane, but there are more equiaxed subgrains (grains) on the X-plane than on the Y-plane. The structure observed on the Y-plane was heterogeneous, which seems to depend on the original grain orientation. For the specimen produced by the 90o die, the banded structure was found to align in a direction varying from 17 ~ 47o from the extrusion direction. However, it is along a direction about 30 ~ 46o from the extrusion direction for the specimen deformed by the 120o die.
The TEM Kikuchi pattern analysis showed that the fraction of high angle boundaries (HABs) with θ >15o was 26% for specimens produced by both dies, but the specimen produced by the 90o die had higher fraction of low angle boundaries (LABs) with θ < 3o. It was noticed that at strain of 4, the distribution of misorientation angles showed large variation in different areas. For example, in one area of the specimen produced by the 90o die, total of 25 grains were measured and all the boundaries were LABs with θ < 6o, while in another area, about 50% of the boundaries were HABs with θ >15o. These results indicate that the deformation structures are quite heterogeneous, which may depend on the crystallographic orientation [9]. The
difference caused by different
crystallographic orientation is often greater
than that due to different die angles. We tend to conclude that at strain of 4, the different die angles, 90o vs. 120o, show no strong effect on the distribution of boundary misorientation angle.
3.3 Microstructure developed at a strain of ~8
The microstructure developed atε~8 is generally more uniform than that developed atε~4. For the specimen produced by the 120o die, the banded structure is still the main feature on the X-plane, which is nearly parallels to the Y-axis, however, some equiaxed subgrains (grains) also present. There seems no significant difference in the microstructure on the X-plane for specimens deformed toε~4 or 8 by the 120o die. On the contrary, for the specimen deformed by the 90o die, the banded structure becomes less evident on the X-plane atε~8. The structure consists of mainly elongated subgrains, which are roughly aligned along the z-axis.
On the Y-plane of the specimen produced by the 90o die, similar to that observed on the X-plane, the banded structure becomes less evident and only barely recognizable in certain areas. Structure consisting of nearly equiaxed subgrains (grains) becomes more popular in the sample. On the contrary, the banded structure is still the dominant feature on the Y-plane of specimens deformed by the 120o die. It is well aligned along a direction about 50 ~ 60o from the extrusion direction, and the dislocation density is still high in the banded structure. Combine the observations on both X- and Y-planes, the main feature of the deformed structure in specimens produced by the 120o die is the banded structure, which is roughly parallel to the shear plane in the die.
At the strain of ~ 8, the deformation structure is quite different for specimens produced by these two dies. However, the different die angles, 90o vs. 120o, show no strong effect on the distribution of boundary misorientation angle. The fraction of high angle boundaries (HABs) with θ > 15o are
4
38% and 35% for specimens produced by the 90o die and 120o die, respectively. The fraction of HABs produced by the ECAE route C is much lower than that produced by the ECAE route A [10], but is higher than that given by the CEC process [11].
4. Conclusions
The important findings of this study can be summarized as follows.
l The observations showed that on the average, the specimen deformed one pass by the 90o die has higher frequency of high angle (>30o) misorientation. The EBSD study also showed that the 90o die results in higher rate of evolution of high angle misorientation at lower strain range (<4).
l At a strain of ~4, the deformed microstructures are quite heterogeneous, which may depend on the local crystallographic orientation. Generally speaking, the microstructure developed by both dies has similar characteristics, banded structure and equiaxed subgrains (grains). However, the 90o die produces higher fraction of equiaxed subgrains (grains).
l At a strain of ~8, the deformed microstructures are more uniform, and the effect of the die angle becomes more pronounced. The 120o die produces mainly banded structure, which are aligned roughly parallel to the shear plane of the die, while with the 90o die, the structure is mainly equiaxed subgrains (grains) and a large fraction of the boundaries has grain boundary fringes.
l Even though there are distinct differences on the microstructure produced by these two dies, the distributions of the boundary misorientation angle are quite similar for both cases. In general, the fraction of HABs increases with increasing strain and it reaches a level of ~38% at strain of ~8.
References
1. R. Z. Valiev, R. K. Islamgaliev, I. V. Alexandrov, Prog. Mater. Sci. 45 (2000) 103.
2. V. M. Segal, Mater. Sci. Eng. A197 (1995) 157.
3. Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto, T. G. Langdon, Scr. Mater. 35 (1996) 143.
4. V. M. Segal, Mater. Sci. Eng. A271 (1999) 322.
5. M. Furukawa, Y. Iwahashi, Z. Horita, M. Nemoto, T. G. Langdon, Mater. Sci. Eng. A257 (1998) 328.
6. Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto and T. G. Langdon, Scripta Mater. 35 (1996) 143.
7. Y. Iwahashi, Z. Horita, M. Nemoto, T. G. Langdon, Acta Mater. 46 (1998) 3317.
8. Q. Liu, J. Appl. Cryst. 27 (1994) 755. 9. D. A. Hughes, N. Hansen, Acta Mater.
45 (1997) 3871.
10. F. J. Humphreys, P. B. Prangnell, J. R. Bowen, A. Gholinia and C. Harris, Phil. Trans. R. Soc. Lond. A357 (1999) 1663.
11. M. Richert, Q. Liu, N. Hansen, Mater. Sci. Eng. A260 (1998) 275.
Figure 1. Schematic illustration of the coordinate associated with the ECAE deformed material.
Z X Y
