- Email: [email protected]
Influence of finish rolling temperature on the microstructure and mechanical properties of Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag alloy sheets
Materials Characterization 139 (2018) 38–48
Contents lists available at ScienceDirect
Materials Characterization journal homepage: www.elsevier.com/locate/matchar
Influence of finish rolling temperature on the microstructure and mechanical properties of Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag alloy sheets
T
⁎
Li Caoa, Chuming Liua, Yonghao Gaoa, , Shunong Jiangb, Yunfeng Liua a b
School of Materials Science and Engineering, Central South University, Changsha 410083, China School of Civil Engineering, Central South University, Changsha 410083, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Magnesium alloy Hot rolling Microstructure Shear zones Anisotropy
Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag (wt%) magnesium sheets were produced with various final rolling temperatures, and their microstructures and mechanical properties were characterized. It was found that low final rolling temperature (LT) resulted in stronger heterogeneity, stronger basal texture and higher strength compared with the high final rolling temperature (HT). Considerable shear zones were observed within both the LT and HT sheets, even though their microstructures, formation mechanisms and mechanical properties were different. Preliminary analysis suggests that the shear zones exert great influence on the mechanical anisotropy of whole sheets.
1. Introduction Magnesium alloys have been attracting attentions from many researchers in recent years, and have become one of the most promising metallic structural materials, especially in the automotive and aerospace industry owing to their low density, high specific strength, high dimensional stability, high damping capacity and recyclability [1–3]. Magnesium alloys have a hexagonal close-packed (HCP) crystal structure with limited number of operative slip systems near room temperature, and the HCP structure also contributes to their poor formability [4,5]. Therefore, plastic processing of magnesium alloys is usually conducted at elevated temperatures, at which additional slip systems become available [6,7]. Hot rolling is a high-efficiency way to produce magnesium alloy sheets, and considerable efforts have been made to investigate the microstructure and mechanical properties of magnesium alloy sheets rolled under various conditions [8–10]. Texture and grain size, which are dependent on rolling temperature, have great impacts on the mechanical properties of magnesium sheets [10–13]. It has been reported that, with decreasing deformation temperature, both basal pole intensity became stronger [10,12] and the split texture peak with a large inclination returned to a single peak [10]. According to the Zener-Hollomon parameter formula, the DRX grain size is reduced with decreasing temperature. And the DRX mechanisms connection with the operated deformation modes are also influenced by temperature [7]. Gledhill [7] systematically investigated the temperature dependence of the operated deformation mode, it was found that the operated deformation mode evolve from twinning, ⟨a⟩
⁎
basal slip and ⟨a + c⟩ dislocation glide at LT, to extensive cross-slip at intermediate temperature, and then to dislocation climb at HT. To conclude, the rolling temperature is a critical parameter for the microstructure and mechanical properties of magnesium alloy. Although numerous studies have been conducted to figure out the relationship between rolling temperature, microstructure and mechanical properties during the rolling process [7,12,14,15], little information about the effect of lowering finish rolling temperature is available, especially in thick plate processing. Combination of hot rolling and warm finish rolling will produce alloy sheets with favorable formability, fine grain size, strong texture and high strain hardening effect. Moreover, the LT finish rolling can be realized using existing facilities, which could reduce production costs. The aim of this study is to understand the relationship between finish rolling temperature and microstructure and mechanical properties by combination of hot rolling and warm rolling, so as to provide a guidance for industrial production. 2. Experimental Procedures 2.1. Materials Preparation High quality Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag (wt%) ingot with 450 mm in diameter was produced by semi-continuous casting method. The solution treatment was performed at 415 °C for 20 h to dissolve the Mg17Al12 phase [16]. Then, the ingot was forged to a true strain of 0.6 at 400 °C to eliminate the cast defects and enhancing the formability of the block in the following rolling process. Blanks with a dimension of
Corresponding author. E-mail addresses: [email protected] (Y. Gao), [email protected] (S. Jiang).
https://doi.org/10.1016/j.matchar.2018.02.014 Received 7 December 2017; Received in revised form 22 January 2018; Accepted 13 February 2018 Available online 15 February 2018 1044-5803/ © 2018 Elsevier Inc. All rights reserved.
Materials Characterization 139 (2018) 38–48
L. Cao et al.
degree for once) and extrapolate wild spikes (once). FEI Quanta-200 SEM equipped with an Oxford energy dispersive X-ray spectrometer (EDS) was applied to study the secondary phase in the alloy. The sample for macro-texture analysis was machined from the middle layer of the sheets, and the test was conducted on Bruker D8 Discover X-ray diffractometer (XRD).
Table 1 Sample designations corresponding to different rolling procedures. Designation
Rolling procedure
AR RH RL
Rolled by 6 passes at 400 °C Finish rolled at 350 °C Finish rolled at 250 °C
2.3. Mechanical Tests 40 mm × 80 mm × 180 mm were cut from the forged billet for subsequent rolling. The hot rolling was conducted on a double roll mill, with a diameter of 460 mm and rolling speed of 0.36 m/s, and the thickness reduction is 14% per rolling pass. The samples were held at 400 °C for 2 h in a resistance furnace, and the roller were preheated to 150 °C prior to rolling. The temperature of samples decreases by ~20 °C per pass. To maintain the processing temperature and make the samples deformed uniformly, the samples were reheated in the resistance furnace for 4 min and reversed after every two passes (i.e. the total thickness reduction between every two successive reheating treatments was ~25%). The samples were air-cooled after 6 passes rolling (designated as AR sheets) to 350 °C or 250 °C within 4 min. Thereafter, finish rolling was conducted with the same thickness reduction. The rolled samples were quenched into water at room temperature immediately after the final pass to preserve the rolling microstructure for further characterization. For convenience, the samples that are finish rolled at 350 °C and 250 °C are denoted as RH and RL, respectively.
For mechanical properties test, dog-bone tensile samples along rolling direction (RD) and transverse direction (TD) with a gauge length of 15 mm, cross-sectional area of 6 mm × 2 mm were cut from the sheets by electric spark wire-cutting. Tensile tests were carried out using an Instron 3369 testing machine at ambient temperature at a cross head speed of 1 mm/min. Triple tests were conducted for each sample and the averaged values were adopted as the mechanical properties. Vickers hardness tests were conducted with a load of 4.9 N and dwelling time of 15 s. 10 indentations were measured for each hardness test and their average was reported here to assure the reliability (Table 1). 3. Results 3.1. Microstructure Fig. 1a and b give the optical microstructure of sample AR. The microstructure is relatively homogeneous, which consists of coarse deformed grains and fine DRX-ed grains in bimodal distribution. In addition, a small amount of AlMn particles with stable refractory phase are randomly distributed in the matrix, which were identified by the EDS analysis, as shown in Fig. 1c and d. The low magnification metallographs of sample RH and RL are exhibited in Fig. 2. It is apparent that microstructure of sample AR changes greatly after the LT finish rolling. Specifically, both samples exhibit striking inhomogeneous deformation microstructure, in which numerous parallel shear zones (along the dashed lines in Fig. 2) distributed symmetrically with respect to the middle plane, and the
2.2. Microstructure Characterization Samples for microstructure and texture analysis were machined from the middle region along the sheet thickness. Leica optical microscope (OM) and FEI Helios Nanolab 600i scanning electron microscope (SEM) equipped with electron backscatter diffraction (EBSD) detector were utilized to investigate the microstructure of the samples. EBSD results were analyzed using HKL Channel 5 software, and the data cleanup procedures including extrapolates zero solution (in moderate
Fig. 1. Optical microstructure of sample AR with different magnification (a) (b), back scattered-electron based SEM micrographs of sample AR (c), and EDS analysis results of the white point in (c) are presented in the table in (d).
39
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 2. Optical micrographs of the as rolled sheets with different finish rolling temperatures: (a), (c) and (e) for the upper, middle and lower regions of sample RH; (b), (d) and (f) for the corresponding regions of sample RL. The magnification of (a–d) are the same as (e–f).
hit rate in the EBSD indexing routine, indicating that the highly strained microstructures formed in the rolling process, especially in the localized shear zones, are only partially recovered/recrystallized. Moreover, the sample RL presents more fraction of zero-solution regions than the sample RH, suggesting that the deformed microstructures are recovered to a higher extent at lower finish rolling temperatures given that the total thickness reductions are the same in both samples. Another distinctive feature of Fig. 4 lies in that the majority of the grains in both samples have got their basal planes aligned in the rolling plane (RD-TD plane). As consistent with the optical micrographs shown in Fig. 3, the inhomogeneous grain size distribution, comprising large deformed grains and fine DRX-ed grains, are also obvious in the EBSD IPF figures (Fig. 4), and the sample RH presents even finer DRX-ed grains due to its higher DRX extent. Two kinds of grain boundaries can be identified in Fig. 4: the normal high angle grain boundaries (HAGB) (presented as black lines) and the low angle grain boundaries (LAGB) (exhibited as white lines).The increase of LAGB from RH to RL(Fig. 4a, b) can represents the increase of dislocation density qualitatively [17]. The grain boundary and crystallographic orientation examination based on EBSD will be presented latter. To understand the effect of finish rolling temperature on the DRX behavior, grain distribution maps of RH and RL are derived from the EBSD data and presented in Fig. 5. A general criterion, i.e. misorientation larger than 7.5°, is adopted in the present study to distinguish the neighboring grains, and the components with misorientation angle 1–7.5° are defined as subgrains. All grains can be classified as deformed, substructured or DRX-ed based on the average intra-granular orientation spread for each individual grain, i.e., those with internal misorientation > 1° are considered as deformed grains (the red grains in Fig. 5), while grains consist of subgrains, whose internal misorientation is under 1° but the misorientation from subgrain to subgrain
normal directions of shear planes incline about 30° from TD to RD. In addition, the density of shear zones decreases when moving from the surface to middle region of the sheets, which might be due to the inhomogeneous distribution of shear strain caused by the friction between the rolls and the surface layer of the sheets. It is understandable that the magnitude of the friction-induced shear strain decreases from the sheet surface to its middle region. To qualitatively establish the correlation between the shear zones evolution vs. shear strain, the microstructure evolution along the sheet thickness was investigated and the corresponding results are exhibited in Fig. 3, in which the optical micrographs of middle, 1/4 thickness and surface regions of sample RH and RL are presented. Generally, the microstructure of the both sheets various from their middle to surface regions. In the middle region, where the shear strain is the least, the microstructure is mainly composed of large deformed grains and fine DRX-ed grains (Fig. 3a and b). When moderate shear strain is applied (1/4 thickness region, Fig. 3c and d), the number fraction of the deformed grains reduced while that of the DRX-ed grains increased. In the surface regions of the sheets, where the severest shear strain is imposed, apparent shear zones are formed, as shown in Fig. 3e and f. Apart from the shear strain, rolling temperature also affects the microstructure characteristics. Specifically, (1) the average grain size of the RL sheet (6.7 μm) is apparently larger than that of the sample RH (14.2 μm); (2) the sample RL features considerable deformation twins, which can hardly be observed in the sample RH. Such differences might be attributed to the larger DRX extent in the sample RH other than in the RL, i.e. higher rolling temperature facilitates the DRX process. The corresponding EBSD inverse pole figure (IPF) maps obtained from longitudinal section of the samples RH and RL are presented in Fig. 4. Obviously, there are some zero-solution regions (24% for RH, 46%for RL) in the IPF figures, which might be attributed to the limited
40
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 3. The optical micrographs of the middle, 1/4 thickness and surface region of sample RH (a, c, e) and RL (b, d, f).
their corresponding boundaries are highlighted with different colors, i.e. the red, green, yellow and white color, respectively, represent the {1012} < 1011> tension twins, {1011} − {1012} double twins, {1011} < 1012> compression twins and {1011} − {1012} double twins. Considering the texture characteristics and applied stress status should have facilitated the compression twinning, it is surprising that only marginal compression twins are observed in the sample RL, which, following Barnett's study [20], might be attributed to the evolution of primary compression twins into the secondary twinning in the subsequent rolling deformation. Also, the deformation twins also affect the formation of shear zones, which will be discussed in the next section.
is above 1°, are classified as substructured components in yellow, the rest are considered as DRX-ed grains in blue. Although both the samples exhibit a mixture of deformed, substructured and DRX-ed grains, obvious differences can still be derived in terms of the fractions of the three categorized grains. While an appreciable fraction (33%) of blue grains are presented in the sample RH, the blue grains can hardly be observed in the sample RL (only occupies ~3%), indicating that higher rolling temperature facilitates the DRX process. On the other hand, more deformed grains may be preserved when the rolling temperature is low, as indicated by the preponderance of red color in sample RL. It is clear that the deformed, substructured and DRX-ed grains are not evenly mixed in the studied samples, rather they exhibit some aggregated characteristics. Specifically, the DRX-ed grains (blue grains) are predominately gathered in the shear zones (as circled by the white ellipse), and only a small fraction is formed surrounding the substructured grains (as highlighted by the white rectangle). It can be speculated that the deformation twins formed in the initial stage of rolling (as confirmed in Fig. 3) may induce significant effect on the microstructure evolution during the subsequent rolling deformation. To figure out this, EBSD Kikuchi band contrast of the sample RL is derived and presented in Fig. 6, in which four types of twin variants are identified based on their characteristic misorientations [18,19] and
3.2. Texture and Mechanical Properties It has been verified that the rolling plane exhibits strong basal texture from Fig. 4. To further figure out the macro-texture of rolled sheets, XRD-based macro-textures of the middle layer of sample AR, RH and RL are given in Fig. 7. In general, the three samples all present strong basal texture, with a maximum pole intensity of 6.14, 7.07 and 9.43, respectively. The pole intensity exhibits an increasing trend with finish rolling temperature decreasing, which is consist with the common trend in magnesium processing [2]. Besides, the texture shape
41
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 4. EBSD IPF maps of as rolled sheets. (a) For RH, (b) for RL. High angle grain boundaries (HAGBs) with misorientation larger than 15° and low angle grain boundaries (LAGBs) with misorientation of 2–15° are represented by black and white lines, respectively.
influence on the microstructure and structural homogeneity of magnesium alloys. As shown in Fig. 9, Ion et al. [15] proposed a deformation behavior map describing the approximate relation between deformation behavior and processing parameters. The fraction of characteristic microstructure (i.e. twinning, recrystallization and shear zones) changes according to the process temperature and strain. Herein, the twinning and shear zones dominate at LT and recrystallization dominate at HT. Locating the parameter ranges of the present work on the deformation map, one can find an excellent coincidence between the proposed model and our experimental results. Sample RH consists of three different parts: deformed grains, DRX-ed grains and macroscopical shear zones (Figs. 3e, 5a), while sample RL is comprised of deformed grains, extensive twins, less DRX-ed grains and more macroscopical shear zones (Figs. 3f, 5b). Due to shear zones only take little fraction in volume in the middle layer of sheets, the XRD-based macrotexture can approximately equal to the texture of homogeneous region. Both sample RH and RL show strong basal texture (Fig. 7b, c), and the intensity of basal texture increase with rolling temperature decreasing. It is well discussed that the orientation of DRX grains are more random than original grains [12,15]. Therefore, the basal texture is weakened with the fraction of DRX grains increase from 3% of RL to 33% of RH. In addition, the critical resolved shear stress (CRSS) of prismatic and pyramidal slip are greatly influenced by temperature, while basal slip and twinning are not sensitive to temperature [6]. Consequently, basal slip take higher fraction in RL compare to RH, which contributes to the formation of stronger basal texture [23]. In conclusion, the increase of basal texture
that basal pole roughly takes the central position of RD-TD plane, which is also widely observed in rolled magnesium alloys [21]. Room temperature tensile strain-stress curves of sample RH and RL along RD and TD are presented in Fig. 8, in which the local parts of the curves are magnified for a clear contrast of the yield strength. In addition, for the convenience of comparison, the ultimate tensile strength (UTS), yield strength (YS) and elongation to failure (Ef) are summarized in Table 2. One can find that sample RH shows better plasticity with Ef up to 18%, while sample RL possesses higher strength with YS along RD up to 285.2 MPa. For sample RH, the YS along RD (231.2 MPa) is 5.9 MPa lower than that along TD (237.1 MPa), while sample RL exhibits a different YS anisotropy, with the YS in RD being 18.4 MPa higher than that in TD. It is widely accepted that mechanical anisotropy is deeply connected with the texture of magnesium alloy [22,23]. However, the convert of yield strength anisotropy in this experiment cannot be explained by the measured macro-texture (Fig.7). Further, we reasonably hypothesis that shear zones play a critical role for mechanical anisotropy, which will be discussed latter. 4. Discussion 4.1. Influence of Finish Rolling Temperature on Microstructure As discussed in previous studies [7,15], with the decrease of temperature, the activated deformation mechanisms change from the HT dislocation climb, to the medium temperature cross-slip, and then to the LT basal slip and twinning, which correspondingly has a significant
42
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 5. Recrystallization grain distribution maps of sample RH (a) and sample RL (b).
15°–20°. The inclination of basal planes of grains in shear zones is widely observed, and commonly attributed to the activation of basal slip or twinning inside [4,15,26]. Ion et al. [15] proposed an opinion that rotation recrystallization (RRX) in shear zones was responsible for the texture inclination and may help to explain the formation process of shear zones in RH. In this study, the formation process of shear zones (i.e. initiation stage, growth stage and completion stage) are captured across the thickness of sheets (Fig. 3), which can testify the connection between shear zones and shear strain. In the light of above research and microstructure analysis, corresponding schematic diagram of the formation process are mapped (Fig. 11). When finish rolling at 350 °C, firstly DRX-ed grains formed in the mantle region of original grains at low strain (i.e. initiation stage, Figs. 3a, 11), and the newly formed recrystallization grains with more random orientation became a soft region. Therefore, during further deformation strain was concentrated in soft region, which broadened the shear zones and connected these regions with others to accommodate more strain (i.e. growth stage Figs. 3c, 11). At completion stage (Figs. 3e, 11), the strain was concentrated inside the shear zones, and mostly accommodated by ⟨a⟩ basal slip due to its lower CRSS [27]. Consequently, the orientation of grain inside the shear zones was changed by specific slip system operation [23]. The macroscopic shear zones with titled basal texture were formed.
intensity is mainly result from the less DRX gains and stronger basal slip. 4.2. Shear Zones in RH 4.2.1. Microstructure In this study, the deformation heterogeneity is aggravated by the strong initial basal texture of sample AR (Fig. 6) and lower finish rolling temperature, which makes the shear zones become the most noticeable feature of the rolled microstructure (Figs. 2, 3). There are extensive shear zones parallelly distributed in both samples (Fig. 2), and the shear zones exhibit different features with finish rolling temperature decreasing (Fig. 3). Many researchers have investigated the shear zones in cold rolling [4,9,24,25]. However, the high temperature shear zones are rarely mentioned [26]. For sample RH in high temperature rolling, the shear zones are deeply connected with DRX behavior (Fig. 3a, c, e). Valle et al. [26] found similar high temperature shear zones and held the view that shear zones originated from general DRX-ed grains, which formed in the mantle region of deformed grains at initial stage of HT deformation, and will be enlarged by increasing strain. To investigate the shear zones in detail, the EBSD-based micro-texture of shear zones (i.e. the regions circled by white ellipse in Fig. 5a) and whole sample RH are calculated and presented in Fig. 10. Obviously, the maximum value of pole intensity of shear zones (9.3) is lower than that of whole sample RH (12.8), which can be attributed to the rather high fraction of DRX-ed grains in shear zones [12,15]. In addition, it is interesting to find that the basal pole of shear zones inclines from ND to RD by
4.2.2. Mechanical Properties In this study, the homogeneous deformation region of both samples
43
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 6. Kikuchi band contrast figures of sample RL (a), the rectangle region in (a) is magnified in (d), detailed analysis of the twin boundary misorientations in (d) are shown in (b–c).
Fig. 7. XRD-based {0002} pole figure of (a) sample AR, (b) sample RH and (c) sample RL.
44
Materials Characterization 139 (2018) 38–48
L. Cao et al.
12a) are almost totally colored by blue, which means that most of the concentrated dislocations inside shear zones are released by DRX. In addition, to investigate the mechanical properties in shear regions in detail, average Schmid factors of basal slip corresponding to the shear zones are calculated and presented in Fig. 13. The average Schmid factors and fraction of the part easy to slip (i.e. Schmid factor larger than 0.35) are summarized in Table 3. It is interesting to note that RD (37%) is almost twice than TD (21%) in terms of easy to slip part. In other words, the shear zones act as easy to slip channels during the tensile test along RD. Specifically, the differences in Schmid factors cause a softening effect along RD compared with TD in shear zones (~44 MPa). On the other side, the Hall-Petch effect [31–33] results in an isotropic strengthening effect by grain size difference between the shear zones and homogeneous deformation region. Considering that the YS can roughly equal to the critical stress for the activation of dislocation movement in polycrystalline materials [34], the yield behavior occurred inside shear zones is anisotropy (i.e. YS along RD is lower than TD by ~44 MPa). It has been well discussed that the overall mechanical properties of heterogeneous materials is decided by all parts [35]. In this study, we reasonable speculated that the mechanical anisotropy of sample RH is attributed to the combination effects of the shear zones (anisotropic) and homogeneous region (isotropic).
Fig. 8. Room temperature tensile curves of sample RH and RL with loading direction parallel to RD or TD. (For example, RH-RD represents the curve for the RD tensile test of sample RH, and the rest designations follow the same manner).
Table 2 Room temperature mechanical properties of sample RH and RL along RD and TD. The standard deviations are in parentheses.
4.3. Shear Zones in RL
Fig. 9. Semi-quantitative diagram summarizing the deformation behavior as a function of temperature and strain [15].
4.3.1. Microstructure As shown in Fig. 3, the formation process of shear zones in sample RL, which are closely related to twins, are slightly different from sample RH. EBSD measurements conducted by Basu et al. [36] on compressed Mg–1Gd alloys with strong basal type starting texture revealed that recrystallization occurred in compression and double twins (i.e. TDRX), and the ⟨c⟩ axis of new recrystallization grains was aligned parallel to the host twin plane. Moreover, Barnett et al. [4] claimed that nucleation of shear zones formed in low temperature rolling originated from double twinning. In addition, the strain induced recrystallization mechanism in warm deformation discussed by Galiyev et al. [37] is also applicable to this study. In the light of the above research, the formation process of the shear zones in RL can be summarized as follows. At initiation stage (Fig. 3b), extensive twins formed under low strain. Herein the basal plane of the compression and double twins inclined by 56°/38° from the compression axis (Fig. 6). After twinning, the soft orientation of twins generated extensive dislocations slip, and high dislocation density motivated continuous dynamic recrystallization (CDRX) inside twins eventually [36]. At growth stage (Fig. 3d), which is similar to that of sample RH, recrystallization grains inside the original twins became the soft region due to more random orientation, and were elongated with increasing strain. At completion stage (Fig. 3f), the recrystallization zones connected with each other, which can accommodate more strain inside. Meanwhile, new equiaxed grains developed at grain and twinning boundaries, and the macroscopic shear zones were formed.
exhibits typical basal texture, with basal pole located at the center of pole figure (Fig. 7). However, the YS of both samples exhibit anisotropy (Fig. 8 & Table 2), which indicate that the anisotropy is associated with inhomogeneous region. To figure out this, the local misorientation map based on EBSD of sample RH and RL were examined (Fig. 12). The color evolution from blue to green and eventually to red represents the increase of dislocation density [28–30]. Both of them exhibit inhomogeneous distribution of dislocation density through the samples. For sample RH (Fig. 12a), dislocation density is low in most regions as colored by blue, and slightly increases in the mantle region of deformation grains and sub-grains boundaries. It is worthwhile to note that DRX-ed grains inside shear zone (i.e. the ellipse region in Figs. 5a,
4.3.2. Mechanical Properties The Vickers hardness presented in Table 4 indicate that the shear zones and homogeneous deformation region in the RL sample also exhibits different mechanical behavior. In detail, the shear zones become “hard” region compared with homogeneous deformation region. This phenomenon can be mainly attributed to two reasons: Hall-Petch effect and work hardening. The grain size inside shear zones and homogeneous region are 1.1 μm and 16.8 μm respectively, rendering higher yielding strength in the shear zones than that in the homogeneous region [32,33]. Opposite to the microstructural characteristics observed in RH (Fig. 12a), a large number of dislocations are accumulated inside and near the shear zones of RL (Fig. 12b), storing large fraction of the deformation energy in the heavily deformed shear zones [38]. The concentrated dislocation formed during low temperature rolling process
Sample
UTS/MPa
YS/MPa
Ef/%
RH-RD RH-TD RL-RD RL-TD
329.7(3.1) 330.5(0.6) 346.9(0.4) 352.6(4.3)
231.3(1.8) 237.1(1.8) 285.2(3.7) 266.8(2.6)
18.7(1.35) 17.7(0.25) 8.8(1.25) 12.4(2.3)
45
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 10. EBSD-based {0002} and {1010} pole figures of sample RH (a–b), and shear zones inside (c–d).
Fig. 11. Schematic diagram of the formation process of shear zones in sample RH.
from 231.3 MPa to 285.2 MPa. Low temperature finish rolling is an effective way to strengthen magnesium alloy sheets. 2. The shear zones formed at different temperature have different formation mechanism, microstructure structures and mechanical properties. For sample RH, the shear zones originated in the general DRX region, and facilitated by concentrated shear strain. The recrystallization grains inside shear zones with inclined basal planes provide an easy way for basal slip especially along RD when elongated, and consequently weaken the yield strength of whole sheet along RD. On the other hand, the shear zones in sample RL originated in the compression and double twins, and also facilitated by concentrated shear strain. The combination of concentrated dislocation and fine grains inside contribute to the high strength of shear zones.
also produced local stress field. Wu et al. [35] proposed that large strain gradient near the domain interface produce back-stress to strength the material during the deformation of heterogeneous materials. In the light of this, we reasonably speculate the presence of back-stress near the shear zones boundaries, and the back-stress is directional due to the anisotropy of shear strain. In this regards, the mechanical anisotropy of sample RL might be attributed to the combined effects of the shear zones (anisotropic) and homogeneous region (isotropic). The mechanical anisotropy of shear zones is hard to figure out, and further research work need to be done. 5. Conclusions 1. After low temperature finish rolling process, the major microstructure features change from deformed grains, general DRX-ed grains and shear zones in sample RH to deformed grains, twins and shear zones in sample RL. In addition, the basal texture intensity and mechanical properties are enhanced with the finish rolling temperature decreased. Herein the yield strength along RD is improved
Further investigations need to give more information on the details of the back-stress produced by the concentrated dislocation in the shear zones of sample RL, for the purpose of giving a deep insight into the mechanical anisotropy of LT finish rolling sheet.
46
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 12. The local misorientation maps of sample RH (a), and RL (b). (The color evolution from blue to red represents the deepening degree of local misorientation). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 13. Schmid factor distribution maps based on basal ⟨a⟩ slip for shear zones along RD (a) and TD (b).
Table 3 The average Schmid factors and the large than 0.35 fraction based on basal ⟨a⟩ slip for shear zones in sample RH relating to the specified directions. Direction
Average Schmid factor
Larger than 0.35
RD of shear zones TD of shear zones
0.27 0.22
37% 21%
Table 4 The Vickers hardness of shear zones and homogeneous region in sample RL. The standard deviations are in parentheses.
47
Sample RL
Shear zones
Matrix region
ND-TD plane ND-RD plane
78.5(1.5) 78.6(2.4)
73.8(1.5) 73.6(2.3)
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Acknowledgements [18]
This work was supported by State Key Laboratory of Powder Metallurgy, Central South University [Grant number 621020097], the National Natural Science Foundation of China [Grant number 51574291].
[19]
[20]
References
[21]
[1] T. Mohri, M. Mabuchi, M. Nakamura, T. Asahina, H. Iwasaki, T. Aizawa, K. Higashi, Microstructural evolution and superplasticity of rolled Mg-9Al-1Zn, Mater. Sci. Eng. A 290 (2000) 139–144. [2] F. Czerwinski, Magnesium and Its Alloys, WILEY (1960). [3] C. Xu, M.Y. Zheng, S.W. Xu, K. Wu, E.D. Wang, S. Kamado, G.J. Wang, X.Y. Lv, Ultra high-strength Mg–Gd–Y–Zn–Zr alloy sheets processed by large-strain hot rolling and ageing, Mater. Sci. Eng. A 547 (2012) 93–98. [4] M.R. Barnett, M.D. Nave, C.J. Bettles, Deformation microstructures and textures of some cold rolled Mg alloys, Mater. Sci. Eng. A 386 (2004) 205–211. [5] C.-j. Li, H.-f. Sun, W.-b. Fang, Effect of extrusion temperatures on microstructures and mechanical properties of Mg-3Zn-0.2Ca-0.5Y alloy, Proc. Eng. 81 (2014) 610–615. [6] G. Bajargan, G. Singh, D. Sivakumar, U. Ramamurty, Effect of temperature and strain rate on the deformation behavior and microstructure of a homogenized AZ31 magnesium alloy, Mater. Sci. Eng. A 579 (2013) 26–34. [7] A. Gledhill, Correlation of plastic deformation and dynamic recrystallization in magnesium alloy ZK60, Acta Mater. 49 (2001) 1199–1207. [8] M.R. Barnett, Z. Keshavarz, M.D. Nave, Microstructural features of rolled Mg-3Al1Zn, Metall. Mater. Trans. A 36 (2005) 1697–1704. [9] M.R. Barnett, N. Stanford, Influence of microstructure on strain distribution in Mg–3Al–1Zn, Scr. Mater. 57 (2007) 1125–1128. [10] X. Huang, K. Suzuki, Y. Chino, M. Mabuchi, Influence of initial texture on rolling and annealing textures of Mg–3Al–1Zn alloy sheets processed by high temperature rolling, J. Alloys Compd. 537 (2012) 80–86. [11] M. Mabuchi, Y. Chino, H. Iwasaki, T. Aizawa, K. Higashi, The grain size and texture dependence of tensile properties in extruded mg-9Al-lZn, Mater. Trans. 42 (2001) 1182–1188. [12] C. Xu, M.Y. Zheng, K. Wu, E.D. Wang, G.H. Fan, S.W. Xu, S. Kamado, X.D. Liu, G.J. Wang, X.Y. Lv, Influence of rolling temperature on the microstructure and mechanical properties of Mg–Gd–Y–Zn–Zr alloy sheets, Mater. Sci. Eng. A 559 (2013) 615–622. [13] W.J. Kim, S.I. Hong, Y.S. Kim, S.H. Min, H.T. Jeong, J.D. Lee, Texture development and its effect on mechanical properties of an AZ61 Mg alloy fabricated by equal channel angular pressing, Acta Mater. 51 (2003) 3293–3307. [14] X. Huang, K. Suzuki, N. Saito, Textures and stretch formability of Mg–6Al–1Zn magnesium alloy sheets rolled at high temperatures up to 793 K, Scr. Mater. 60 (2009) 651–654. [15] S.E. Ion, F.J. Humphreys, S.H. White, Dynamic recrystallisation and the development of microstructure during the high temperature deformation of magnesium, Acta Metall. 30 (1982) 1909–1919. [16] J.M. Zhang, Z.H. Wang, B.L. Jiang, Z.S. Chen, L.B. Niu, Influence of annealing treatment on precipitation of Î2-Mg17Al12 phase of AZ80 magnesium alloy, Adv. Mater. Res. 476-478 (2012) 55–58. [17] P. Molnár, A. Jäger, P. Lejček, The role of low-angle grain boundaries in multi-
[22] [23] [24]
[25]
[26] [27] [28]
[29]
[30]
[31] [32] [33]
[34]
[35] [36] [37] [38]
48
temperature equal channel angular pressing of Mg–3Al–1Zn alloy, J. Mater. Sci. 47 (2011) 3265–3271. J. Koike, Enhanced deformation mechanisms by anisotropic plasticity in polycrystalline Mg alloys at room temperature, Metall. Mater. Trans. A 36 (2005) 1689–1696. S.W. Xu, S. Kamado, N. Matsumoto, T. Honma, Y. Kojima, Recrystallization mechanism of as-cast AZ91 magnesium alloy during hot compressive deformation, Mater. Sci. Eng. A 527 (2009) 52–60. M.R. Barnett, Z. Keshavarz, A.G. Beer, X. Ma, Non-Schmid behaviour during secondary twinning in a polycrystalline magnesium alloy, Acta Mater. 56 (2008) 5–15. J.L. Haughton, The technology of magnesium and its alloys, Nature 148 (1941) 67–68. H. Wang, P.D. Wu, M.A. Gharghouri, Effects of basal texture on mechanical behaviour of magnesium alloy AZ31B sheet, Mater. Sci. Eng. A 527 (2010) 3588–3594. Y.N. Wang, J.C. Huang, Texture analysis in hexagonal materials, Mater. Chem. Phys. 81 (2003) 11–26. S.W. Xu, K. Oh-ishi, S. Kamado, T. Homma, Twins, recrystallization and texture evolution of a Mg–5.99Zn–1.76Ca–0.35Mn (wt.%) alloy during indirect extrusion process, Scr. Mater. 65 (2011) 875–878. N. Stanford, M.R. Barnett, The origin of “Rare Earth” texture development in extruded Mg-based alloys and its effect on tensile ductility, Mater. Sci. Eng. A 496 (2008) 399–408. J.A.D. Valle, M.T. Pérez-Prado, O.A. Ruano, Texture evolution during large-strain hot rolling of the Mg AZ61 alloy, Mater. Sci. Eng. A 355 (2003) 68–78. M.H. Yoo, Slip, twinning, and fracture in hexagonal close-packed metals, Metall. Trans. A. 12 (1981) 409–418. Z. Wang, Z. Chen, C. Zhan, L. Kuang, J. Shao, R. Wang, C. Liu, Quasi-static and dynamic forced shear deformation behaviors of Ti-5Mo-5V-8Cr-3Al alloy, Mater. Sci. Eng. A 691 (2017) 51–59. J.L. Sun, P.W. Trimby, F.K. Yan, X.Z. Liao, N.R. Tao, J.T. Wang, Shear banding in commercial pure titanium deformed by dynamic compression, Acta Mater. 79 (2014) 47–58. H. Li, E. Hsu, J. Szpunar, H. Utsunomiya, T. Sakai, Deformation mechanism and texture and microstructure evolution during high-speed rolling of AZ31B Mg sheets, J. Mater. Sci. 43 (2008) 7148–7156. N.J. Petch, The cleavage strength of polycrystals, J. Iron Steel Inst. 174 (1953) 25–28. N. Hansen, Hall–Petch relation and boundary strengthening, Scr. Mater. 51 (2004) 801–806. S.M. He, X.Q. Zeng, L.M. Peng, X. Gao, J.F. Nie, W.J. Ding, Microstructure and strengthening mechanism of high strength Mg–10Gd–2Y–0.5Zr alloy, J. Alloys Compd. 427 (2007) 316–323. A.S. Khan, A. Pandey, T. Gnäupel-Herold, R.K. Mishra, Mechanical response and texture evolution of AZ31 alloy at large strains for different strain rates and temperatures, Int. J. Plast. 27 (2011) 688–706. X. Wu, Y. Zhu, Heterogeneous materials: a new class of materials with unprecedented mechanical properties, Math. Res. Lett. (2017) 1–6. I. Basu, T. Al-Samman, Twin recrystallization mechanisms in magnesium-rare earth alloys, Acta Mater. 96 (2015) 111–132. A.M. Galiyev, R.O. Kaibyshev, G. Gottstein, Grain Refinement of ZK60 Magnesium Alloy During Low Temperature Deformation, Magnesium Technol. (2002) 181–185. F. Guo, D. Zhang, X. Yang, L. Jiang, S. Chai, F. Pan, Influence of rolling speed on microstructure and mechanical properties of AZ31 Mg alloy rolled by large strain hot rolling, Mater. Sci. Eng. A 607 (2014) 383–389.
Contents lists available at ScienceDirect
Materials Characterization journal homepage: www.elsevier.com/locate/matchar
Influence of finish rolling temperature on the microstructure and mechanical properties of Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag alloy sheets
T
⁎
Li Caoa, Chuming Liua, Yonghao Gaoa, , Shunong Jiangb, Yunfeng Liua a b
School of Materials Science and Engineering, Central South University, Changsha 410083, China School of Civil Engineering, Central South University, Changsha 410083, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Magnesium alloy Hot rolling Microstructure Shear zones Anisotropy
Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag (wt%) magnesium sheets were produced with various final rolling temperatures, and their microstructures and mechanical properties were characterized. It was found that low final rolling temperature (LT) resulted in stronger heterogeneity, stronger basal texture and higher strength compared with the high final rolling temperature (HT). Considerable shear zones were observed within both the LT and HT sheets, even though their microstructures, formation mechanisms and mechanical properties were different. Preliminary analysis suggests that the shear zones exert great influence on the mechanical anisotropy of whole sheets.
1. Introduction Magnesium alloys have been attracting attentions from many researchers in recent years, and have become one of the most promising metallic structural materials, especially in the automotive and aerospace industry owing to their low density, high specific strength, high dimensional stability, high damping capacity and recyclability [1–3]. Magnesium alloys have a hexagonal close-packed (HCP) crystal structure with limited number of operative slip systems near room temperature, and the HCP structure also contributes to their poor formability [4,5]. Therefore, plastic processing of magnesium alloys is usually conducted at elevated temperatures, at which additional slip systems become available [6,7]. Hot rolling is a high-efficiency way to produce magnesium alloy sheets, and considerable efforts have been made to investigate the microstructure and mechanical properties of magnesium alloy sheets rolled under various conditions [8–10]. Texture and grain size, which are dependent on rolling temperature, have great impacts on the mechanical properties of magnesium sheets [10–13]. It has been reported that, with decreasing deformation temperature, both basal pole intensity became stronger [10,12] and the split texture peak with a large inclination returned to a single peak [10]. According to the Zener-Hollomon parameter formula, the DRX grain size is reduced with decreasing temperature. And the DRX mechanisms connection with the operated deformation modes are also influenced by temperature [7]. Gledhill [7] systematically investigated the temperature dependence of the operated deformation mode, it was found that the operated deformation mode evolve from twinning, ⟨a⟩
⁎
basal slip and ⟨a + c⟩ dislocation glide at LT, to extensive cross-slip at intermediate temperature, and then to dislocation climb at HT. To conclude, the rolling temperature is a critical parameter for the microstructure and mechanical properties of magnesium alloy. Although numerous studies have been conducted to figure out the relationship between rolling temperature, microstructure and mechanical properties during the rolling process [7,12,14,15], little information about the effect of lowering finish rolling temperature is available, especially in thick plate processing. Combination of hot rolling and warm finish rolling will produce alloy sheets with favorable formability, fine grain size, strong texture and high strain hardening effect. Moreover, the LT finish rolling can be realized using existing facilities, which could reduce production costs. The aim of this study is to understand the relationship between finish rolling temperature and microstructure and mechanical properties by combination of hot rolling and warm rolling, so as to provide a guidance for industrial production. 2. Experimental Procedures 2.1. Materials Preparation High quality Mg-8.5Al-0.5Zn-0.2Mn-0.15Ag (wt%) ingot with 450 mm in diameter was produced by semi-continuous casting method. The solution treatment was performed at 415 °C for 20 h to dissolve the Mg17Al12 phase [16]. Then, the ingot was forged to a true strain of 0.6 at 400 °C to eliminate the cast defects and enhancing the formability of the block in the following rolling process. Blanks with a dimension of
Corresponding author. E-mail addresses: [email protected] (Y. Gao), [email protected] (S. Jiang).
https://doi.org/10.1016/j.matchar.2018.02.014 Received 7 December 2017; Received in revised form 22 January 2018; Accepted 13 February 2018 Available online 15 February 2018 1044-5803/ © 2018 Elsevier Inc. All rights reserved.
Materials Characterization 139 (2018) 38–48
L. Cao et al.
degree for once) and extrapolate wild spikes (once). FEI Quanta-200 SEM equipped with an Oxford energy dispersive X-ray spectrometer (EDS) was applied to study the secondary phase in the alloy. The sample for macro-texture analysis was machined from the middle layer of the sheets, and the test was conducted on Bruker D8 Discover X-ray diffractometer (XRD).
Table 1 Sample designations corresponding to different rolling procedures. Designation
Rolling procedure
AR RH RL
Rolled by 6 passes at 400 °C Finish rolled at 350 °C Finish rolled at 250 °C
2.3. Mechanical Tests 40 mm × 80 mm × 180 mm were cut from the forged billet for subsequent rolling. The hot rolling was conducted on a double roll mill, with a diameter of 460 mm and rolling speed of 0.36 m/s, and the thickness reduction is 14% per rolling pass. The samples were held at 400 °C for 2 h in a resistance furnace, and the roller were preheated to 150 °C prior to rolling. The temperature of samples decreases by ~20 °C per pass. To maintain the processing temperature and make the samples deformed uniformly, the samples were reheated in the resistance furnace for 4 min and reversed after every two passes (i.e. the total thickness reduction between every two successive reheating treatments was ~25%). The samples were air-cooled after 6 passes rolling (designated as AR sheets) to 350 °C or 250 °C within 4 min. Thereafter, finish rolling was conducted with the same thickness reduction. The rolled samples were quenched into water at room temperature immediately after the final pass to preserve the rolling microstructure for further characterization. For convenience, the samples that are finish rolled at 350 °C and 250 °C are denoted as RH and RL, respectively.
For mechanical properties test, dog-bone tensile samples along rolling direction (RD) and transverse direction (TD) with a gauge length of 15 mm, cross-sectional area of 6 mm × 2 mm were cut from the sheets by electric spark wire-cutting. Tensile tests were carried out using an Instron 3369 testing machine at ambient temperature at a cross head speed of 1 mm/min. Triple tests were conducted for each sample and the averaged values were adopted as the mechanical properties. Vickers hardness tests were conducted with a load of 4.9 N and dwelling time of 15 s. 10 indentations were measured for each hardness test and their average was reported here to assure the reliability (Table 1). 3. Results 3.1. Microstructure Fig. 1a and b give the optical microstructure of sample AR. The microstructure is relatively homogeneous, which consists of coarse deformed grains and fine DRX-ed grains in bimodal distribution. In addition, a small amount of AlMn particles with stable refractory phase are randomly distributed in the matrix, which were identified by the EDS analysis, as shown in Fig. 1c and d. The low magnification metallographs of sample RH and RL are exhibited in Fig. 2. It is apparent that microstructure of sample AR changes greatly after the LT finish rolling. Specifically, both samples exhibit striking inhomogeneous deformation microstructure, in which numerous parallel shear zones (along the dashed lines in Fig. 2) distributed symmetrically with respect to the middle plane, and the
2.2. Microstructure Characterization Samples for microstructure and texture analysis were machined from the middle region along the sheet thickness. Leica optical microscope (OM) and FEI Helios Nanolab 600i scanning electron microscope (SEM) equipped with electron backscatter diffraction (EBSD) detector were utilized to investigate the microstructure of the samples. EBSD results were analyzed using HKL Channel 5 software, and the data cleanup procedures including extrapolates zero solution (in moderate
Fig. 1. Optical microstructure of sample AR with different magnification (a) (b), back scattered-electron based SEM micrographs of sample AR (c), and EDS analysis results of the white point in (c) are presented in the table in (d).
39
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 2. Optical micrographs of the as rolled sheets with different finish rolling temperatures: (a), (c) and (e) for the upper, middle and lower regions of sample RH; (b), (d) and (f) for the corresponding regions of sample RL. The magnification of (a–d) are the same as (e–f).
hit rate in the EBSD indexing routine, indicating that the highly strained microstructures formed in the rolling process, especially in the localized shear zones, are only partially recovered/recrystallized. Moreover, the sample RL presents more fraction of zero-solution regions than the sample RH, suggesting that the deformed microstructures are recovered to a higher extent at lower finish rolling temperatures given that the total thickness reductions are the same in both samples. Another distinctive feature of Fig. 4 lies in that the majority of the grains in both samples have got their basal planes aligned in the rolling plane (RD-TD plane). As consistent with the optical micrographs shown in Fig. 3, the inhomogeneous grain size distribution, comprising large deformed grains and fine DRX-ed grains, are also obvious in the EBSD IPF figures (Fig. 4), and the sample RH presents even finer DRX-ed grains due to its higher DRX extent. Two kinds of grain boundaries can be identified in Fig. 4: the normal high angle grain boundaries (HAGB) (presented as black lines) and the low angle grain boundaries (LAGB) (exhibited as white lines).The increase of LAGB from RH to RL(Fig. 4a, b) can represents the increase of dislocation density qualitatively [17]. The grain boundary and crystallographic orientation examination based on EBSD will be presented latter. To understand the effect of finish rolling temperature on the DRX behavior, grain distribution maps of RH and RL are derived from the EBSD data and presented in Fig. 5. A general criterion, i.e. misorientation larger than 7.5°, is adopted in the present study to distinguish the neighboring grains, and the components with misorientation angle 1–7.5° are defined as subgrains. All grains can be classified as deformed, substructured or DRX-ed based on the average intra-granular orientation spread for each individual grain, i.e., those with internal misorientation > 1° are considered as deformed grains (the red grains in Fig. 5), while grains consist of subgrains, whose internal misorientation is under 1° but the misorientation from subgrain to subgrain
normal directions of shear planes incline about 30° from TD to RD. In addition, the density of shear zones decreases when moving from the surface to middle region of the sheets, which might be due to the inhomogeneous distribution of shear strain caused by the friction between the rolls and the surface layer of the sheets. It is understandable that the magnitude of the friction-induced shear strain decreases from the sheet surface to its middle region. To qualitatively establish the correlation between the shear zones evolution vs. shear strain, the microstructure evolution along the sheet thickness was investigated and the corresponding results are exhibited in Fig. 3, in which the optical micrographs of middle, 1/4 thickness and surface regions of sample RH and RL are presented. Generally, the microstructure of the both sheets various from their middle to surface regions. In the middle region, where the shear strain is the least, the microstructure is mainly composed of large deformed grains and fine DRX-ed grains (Fig. 3a and b). When moderate shear strain is applied (1/4 thickness region, Fig. 3c and d), the number fraction of the deformed grains reduced while that of the DRX-ed grains increased. In the surface regions of the sheets, where the severest shear strain is imposed, apparent shear zones are formed, as shown in Fig. 3e and f. Apart from the shear strain, rolling temperature also affects the microstructure characteristics. Specifically, (1) the average grain size of the RL sheet (6.7 μm) is apparently larger than that of the sample RH (14.2 μm); (2) the sample RL features considerable deformation twins, which can hardly be observed in the sample RH. Such differences might be attributed to the larger DRX extent in the sample RH other than in the RL, i.e. higher rolling temperature facilitates the DRX process. The corresponding EBSD inverse pole figure (IPF) maps obtained from longitudinal section of the samples RH and RL are presented in Fig. 4. Obviously, there are some zero-solution regions (24% for RH, 46%for RL) in the IPF figures, which might be attributed to the limited
40
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 3. The optical micrographs of the middle, 1/4 thickness and surface region of sample RH (a, c, e) and RL (b, d, f).
their corresponding boundaries are highlighted with different colors, i.e. the red, green, yellow and white color, respectively, represent the {1012} < 1011> tension twins, {1011} − {1012} double twins, {1011} < 1012> compression twins and {1011} − {1012} double twins. Considering the texture characteristics and applied stress status should have facilitated the compression twinning, it is surprising that only marginal compression twins are observed in the sample RL, which, following Barnett's study [20], might be attributed to the evolution of primary compression twins into the secondary twinning in the subsequent rolling deformation. Also, the deformation twins also affect the formation of shear zones, which will be discussed in the next section.
is above 1°, are classified as substructured components in yellow, the rest are considered as DRX-ed grains in blue. Although both the samples exhibit a mixture of deformed, substructured and DRX-ed grains, obvious differences can still be derived in terms of the fractions of the three categorized grains. While an appreciable fraction (33%) of blue grains are presented in the sample RH, the blue grains can hardly be observed in the sample RL (only occupies ~3%), indicating that higher rolling temperature facilitates the DRX process. On the other hand, more deformed grains may be preserved when the rolling temperature is low, as indicated by the preponderance of red color in sample RL. It is clear that the deformed, substructured and DRX-ed grains are not evenly mixed in the studied samples, rather they exhibit some aggregated characteristics. Specifically, the DRX-ed grains (blue grains) are predominately gathered in the shear zones (as circled by the white ellipse), and only a small fraction is formed surrounding the substructured grains (as highlighted by the white rectangle). It can be speculated that the deformation twins formed in the initial stage of rolling (as confirmed in Fig. 3) may induce significant effect on the microstructure evolution during the subsequent rolling deformation. To figure out this, EBSD Kikuchi band contrast of the sample RL is derived and presented in Fig. 6, in which four types of twin variants are identified based on their characteristic misorientations [18,19] and
3.2. Texture and Mechanical Properties It has been verified that the rolling plane exhibits strong basal texture from Fig. 4. To further figure out the macro-texture of rolled sheets, XRD-based macro-textures of the middle layer of sample AR, RH and RL are given in Fig. 7. In general, the three samples all present strong basal texture, with a maximum pole intensity of 6.14, 7.07 and 9.43, respectively. The pole intensity exhibits an increasing trend with finish rolling temperature decreasing, which is consist with the common trend in magnesium processing [2]. Besides, the texture shape
41
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 4. EBSD IPF maps of as rolled sheets. (a) For RH, (b) for RL. High angle grain boundaries (HAGBs) with misorientation larger than 15° and low angle grain boundaries (LAGBs) with misorientation of 2–15° are represented by black and white lines, respectively.
influence on the microstructure and structural homogeneity of magnesium alloys. As shown in Fig. 9, Ion et al. [15] proposed a deformation behavior map describing the approximate relation between deformation behavior and processing parameters. The fraction of characteristic microstructure (i.e. twinning, recrystallization and shear zones) changes according to the process temperature and strain. Herein, the twinning and shear zones dominate at LT and recrystallization dominate at HT. Locating the parameter ranges of the present work on the deformation map, one can find an excellent coincidence between the proposed model and our experimental results. Sample RH consists of three different parts: deformed grains, DRX-ed grains and macroscopical shear zones (Figs. 3e, 5a), while sample RL is comprised of deformed grains, extensive twins, less DRX-ed grains and more macroscopical shear zones (Figs. 3f, 5b). Due to shear zones only take little fraction in volume in the middle layer of sheets, the XRD-based macrotexture can approximately equal to the texture of homogeneous region. Both sample RH and RL show strong basal texture (Fig. 7b, c), and the intensity of basal texture increase with rolling temperature decreasing. It is well discussed that the orientation of DRX grains are more random than original grains [12,15]. Therefore, the basal texture is weakened with the fraction of DRX grains increase from 3% of RL to 33% of RH. In addition, the critical resolved shear stress (CRSS) of prismatic and pyramidal slip are greatly influenced by temperature, while basal slip and twinning are not sensitive to temperature [6]. Consequently, basal slip take higher fraction in RL compare to RH, which contributes to the formation of stronger basal texture [23]. In conclusion, the increase of basal texture
that basal pole roughly takes the central position of RD-TD plane, which is also widely observed in rolled magnesium alloys [21]. Room temperature tensile strain-stress curves of sample RH and RL along RD and TD are presented in Fig. 8, in which the local parts of the curves are magnified for a clear contrast of the yield strength. In addition, for the convenience of comparison, the ultimate tensile strength (UTS), yield strength (YS) and elongation to failure (Ef) are summarized in Table 2. One can find that sample RH shows better plasticity with Ef up to 18%, while sample RL possesses higher strength with YS along RD up to 285.2 MPa. For sample RH, the YS along RD (231.2 MPa) is 5.9 MPa lower than that along TD (237.1 MPa), while sample RL exhibits a different YS anisotropy, with the YS in RD being 18.4 MPa higher than that in TD. It is widely accepted that mechanical anisotropy is deeply connected with the texture of magnesium alloy [22,23]. However, the convert of yield strength anisotropy in this experiment cannot be explained by the measured macro-texture (Fig.7). Further, we reasonably hypothesis that shear zones play a critical role for mechanical anisotropy, which will be discussed latter. 4. Discussion 4.1. Influence of Finish Rolling Temperature on Microstructure As discussed in previous studies [7,15], with the decrease of temperature, the activated deformation mechanisms change from the HT dislocation climb, to the medium temperature cross-slip, and then to the LT basal slip and twinning, which correspondingly has a significant
42
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 5. Recrystallization grain distribution maps of sample RH (a) and sample RL (b).
15°–20°. The inclination of basal planes of grains in shear zones is widely observed, and commonly attributed to the activation of basal slip or twinning inside [4,15,26]. Ion et al. [15] proposed an opinion that rotation recrystallization (RRX) in shear zones was responsible for the texture inclination and may help to explain the formation process of shear zones in RH. In this study, the formation process of shear zones (i.e. initiation stage, growth stage and completion stage) are captured across the thickness of sheets (Fig. 3), which can testify the connection between shear zones and shear strain. In the light of above research and microstructure analysis, corresponding schematic diagram of the formation process are mapped (Fig. 11). When finish rolling at 350 °C, firstly DRX-ed grains formed in the mantle region of original grains at low strain (i.e. initiation stage, Figs. 3a, 11), and the newly formed recrystallization grains with more random orientation became a soft region. Therefore, during further deformation strain was concentrated in soft region, which broadened the shear zones and connected these regions with others to accommodate more strain (i.e. growth stage Figs. 3c, 11). At completion stage (Figs. 3e, 11), the strain was concentrated inside the shear zones, and mostly accommodated by ⟨a⟩ basal slip due to its lower CRSS [27]. Consequently, the orientation of grain inside the shear zones was changed by specific slip system operation [23]. The macroscopic shear zones with titled basal texture were formed.
intensity is mainly result from the less DRX gains and stronger basal slip. 4.2. Shear Zones in RH 4.2.1. Microstructure In this study, the deformation heterogeneity is aggravated by the strong initial basal texture of sample AR (Fig. 6) and lower finish rolling temperature, which makes the shear zones become the most noticeable feature of the rolled microstructure (Figs. 2, 3). There are extensive shear zones parallelly distributed in both samples (Fig. 2), and the shear zones exhibit different features with finish rolling temperature decreasing (Fig. 3). Many researchers have investigated the shear zones in cold rolling [4,9,24,25]. However, the high temperature shear zones are rarely mentioned [26]. For sample RH in high temperature rolling, the shear zones are deeply connected with DRX behavior (Fig. 3a, c, e). Valle et al. [26] found similar high temperature shear zones and held the view that shear zones originated from general DRX-ed grains, which formed in the mantle region of deformed grains at initial stage of HT deformation, and will be enlarged by increasing strain. To investigate the shear zones in detail, the EBSD-based micro-texture of shear zones (i.e. the regions circled by white ellipse in Fig. 5a) and whole sample RH are calculated and presented in Fig. 10. Obviously, the maximum value of pole intensity of shear zones (9.3) is lower than that of whole sample RH (12.8), which can be attributed to the rather high fraction of DRX-ed grains in shear zones [12,15]. In addition, it is interesting to find that the basal pole of shear zones inclines from ND to RD by
4.2.2. Mechanical Properties In this study, the homogeneous deformation region of both samples
43
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 6. Kikuchi band contrast figures of sample RL (a), the rectangle region in (a) is magnified in (d), detailed analysis of the twin boundary misorientations in (d) are shown in (b–c).
Fig. 7. XRD-based {0002} pole figure of (a) sample AR, (b) sample RH and (c) sample RL.
44
Materials Characterization 139 (2018) 38–48
L. Cao et al.
12a) are almost totally colored by blue, which means that most of the concentrated dislocations inside shear zones are released by DRX. In addition, to investigate the mechanical properties in shear regions in detail, average Schmid factors of basal slip corresponding to the shear zones are calculated and presented in Fig. 13. The average Schmid factors and fraction of the part easy to slip (i.e. Schmid factor larger than 0.35) are summarized in Table 3. It is interesting to note that RD (37%) is almost twice than TD (21%) in terms of easy to slip part. In other words, the shear zones act as easy to slip channels during the tensile test along RD. Specifically, the differences in Schmid factors cause a softening effect along RD compared with TD in shear zones (~44 MPa). On the other side, the Hall-Petch effect [31–33] results in an isotropic strengthening effect by grain size difference between the shear zones and homogeneous deformation region. Considering that the YS can roughly equal to the critical stress for the activation of dislocation movement in polycrystalline materials [34], the yield behavior occurred inside shear zones is anisotropy (i.e. YS along RD is lower than TD by ~44 MPa). It has been well discussed that the overall mechanical properties of heterogeneous materials is decided by all parts [35]. In this study, we reasonable speculated that the mechanical anisotropy of sample RH is attributed to the combination effects of the shear zones (anisotropic) and homogeneous region (isotropic).
Fig. 8. Room temperature tensile curves of sample RH and RL with loading direction parallel to RD or TD. (For example, RH-RD represents the curve for the RD tensile test of sample RH, and the rest designations follow the same manner).
Table 2 Room temperature mechanical properties of sample RH and RL along RD and TD. The standard deviations are in parentheses.
4.3. Shear Zones in RL
Fig. 9. Semi-quantitative diagram summarizing the deformation behavior as a function of temperature and strain [15].
4.3.1. Microstructure As shown in Fig. 3, the formation process of shear zones in sample RL, which are closely related to twins, are slightly different from sample RH. EBSD measurements conducted by Basu et al. [36] on compressed Mg–1Gd alloys with strong basal type starting texture revealed that recrystallization occurred in compression and double twins (i.e. TDRX), and the ⟨c⟩ axis of new recrystallization grains was aligned parallel to the host twin plane. Moreover, Barnett et al. [4] claimed that nucleation of shear zones formed in low temperature rolling originated from double twinning. In addition, the strain induced recrystallization mechanism in warm deformation discussed by Galiyev et al. [37] is also applicable to this study. In the light of the above research, the formation process of the shear zones in RL can be summarized as follows. At initiation stage (Fig. 3b), extensive twins formed under low strain. Herein the basal plane of the compression and double twins inclined by 56°/38° from the compression axis (Fig. 6). After twinning, the soft orientation of twins generated extensive dislocations slip, and high dislocation density motivated continuous dynamic recrystallization (CDRX) inside twins eventually [36]. At growth stage (Fig. 3d), which is similar to that of sample RH, recrystallization grains inside the original twins became the soft region due to more random orientation, and were elongated with increasing strain. At completion stage (Fig. 3f), the recrystallization zones connected with each other, which can accommodate more strain inside. Meanwhile, new equiaxed grains developed at grain and twinning boundaries, and the macroscopic shear zones were formed.
exhibits typical basal texture, with basal pole located at the center of pole figure (Fig. 7). However, the YS of both samples exhibit anisotropy (Fig. 8 & Table 2), which indicate that the anisotropy is associated with inhomogeneous region. To figure out this, the local misorientation map based on EBSD of sample RH and RL were examined (Fig. 12). The color evolution from blue to green and eventually to red represents the increase of dislocation density [28–30]. Both of them exhibit inhomogeneous distribution of dislocation density through the samples. For sample RH (Fig. 12a), dislocation density is low in most regions as colored by blue, and slightly increases in the mantle region of deformation grains and sub-grains boundaries. It is worthwhile to note that DRX-ed grains inside shear zone (i.e. the ellipse region in Figs. 5a,
4.3.2. Mechanical Properties The Vickers hardness presented in Table 4 indicate that the shear zones and homogeneous deformation region in the RL sample also exhibits different mechanical behavior. In detail, the shear zones become “hard” region compared with homogeneous deformation region. This phenomenon can be mainly attributed to two reasons: Hall-Petch effect and work hardening. The grain size inside shear zones and homogeneous region are 1.1 μm and 16.8 μm respectively, rendering higher yielding strength in the shear zones than that in the homogeneous region [32,33]. Opposite to the microstructural characteristics observed in RH (Fig. 12a), a large number of dislocations are accumulated inside and near the shear zones of RL (Fig. 12b), storing large fraction of the deformation energy in the heavily deformed shear zones [38]. The concentrated dislocation formed during low temperature rolling process
Sample
UTS/MPa
YS/MPa
Ef/%
RH-RD RH-TD RL-RD RL-TD
329.7(3.1) 330.5(0.6) 346.9(0.4) 352.6(4.3)
231.3(1.8) 237.1(1.8) 285.2(3.7) 266.8(2.6)
18.7(1.35) 17.7(0.25) 8.8(1.25) 12.4(2.3)
45
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 10. EBSD-based {0002} and {1010} pole figures of sample RH (a–b), and shear zones inside (c–d).
Fig. 11. Schematic diagram of the formation process of shear zones in sample RH.
from 231.3 MPa to 285.2 MPa. Low temperature finish rolling is an effective way to strengthen magnesium alloy sheets. 2. The shear zones formed at different temperature have different formation mechanism, microstructure structures and mechanical properties. For sample RH, the shear zones originated in the general DRX region, and facilitated by concentrated shear strain. The recrystallization grains inside shear zones with inclined basal planes provide an easy way for basal slip especially along RD when elongated, and consequently weaken the yield strength of whole sheet along RD. On the other hand, the shear zones in sample RL originated in the compression and double twins, and also facilitated by concentrated shear strain. The combination of concentrated dislocation and fine grains inside contribute to the high strength of shear zones.
also produced local stress field. Wu et al. [35] proposed that large strain gradient near the domain interface produce back-stress to strength the material during the deformation of heterogeneous materials. In the light of this, we reasonably speculate the presence of back-stress near the shear zones boundaries, and the back-stress is directional due to the anisotropy of shear strain. In this regards, the mechanical anisotropy of sample RL might be attributed to the combined effects of the shear zones (anisotropic) and homogeneous region (isotropic). The mechanical anisotropy of shear zones is hard to figure out, and further research work need to be done. 5. Conclusions 1. After low temperature finish rolling process, the major microstructure features change from deformed grains, general DRX-ed grains and shear zones in sample RH to deformed grains, twins and shear zones in sample RL. In addition, the basal texture intensity and mechanical properties are enhanced with the finish rolling temperature decreased. Herein the yield strength along RD is improved
Further investigations need to give more information on the details of the back-stress produced by the concentrated dislocation in the shear zones of sample RL, for the purpose of giving a deep insight into the mechanical anisotropy of LT finish rolling sheet.
46
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Fig. 12. The local misorientation maps of sample RH (a), and RL (b). (The color evolution from blue to red represents the deepening degree of local misorientation). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 13. Schmid factor distribution maps based on basal ⟨a⟩ slip for shear zones along RD (a) and TD (b).
Table 3 The average Schmid factors and the large than 0.35 fraction based on basal ⟨a⟩ slip for shear zones in sample RH relating to the specified directions. Direction
Average Schmid factor
Larger than 0.35
RD of shear zones TD of shear zones
0.27 0.22
37% 21%
Table 4 The Vickers hardness of shear zones and homogeneous region in sample RL. The standard deviations are in parentheses.
47
Sample RL
Shear zones
Matrix region
ND-TD plane ND-RD plane
78.5(1.5) 78.6(2.4)
73.8(1.5) 73.6(2.3)
Materials Characterization 139 (2018) 38–48
L. Cao et al.
Acknowledgements [18]
This work was supported by State Key Laboratory of Powder Metallurgy, Central South University [Grant number 621020097], the National Natural Science Foundation of China [Grant number 51574291].
[19]
[20]
References
[21]
[1] T. Mohri, M. Mabuchi, M. Nakamura, T. Asahina, H. Iwasaki, T. Aizawa, K. Higashi, Microstructural evolution and superplasticity of rolled Mg-9Al-1Zn, Mater. Sci. Eng. A 290 (2000) 139–144. [2] F. Czerwinski, Magnesium and Its Alloys, WILEY (1960). [3] C. Xu, M.Y. Zheng, S.W. Xu, K. Wu, E.D. Wang, S. Kamado, G.J. Wang, X.Y. Lv, Ultra high-strength Mg–Gd–Y–Zn–Zr alloy sheets processed by large-strain hot rolling and ageing, Mater. Sci. Eng. A 547 (2012) 93–98. [4] M.R. Barnett, M.D. Nave, C.J. Bettles, Deformation microstructures and textures of some cold rolled Mg alloys, Mater. Sci. Eng. A 386 (2004) 205–211. [5] C.-j. Li, H.-f. Sun, W.-b. Fang, Effect of extrusion temperatures on microstructures and mechanical properties of Mg-3Zn-0.2Ca-0.5Y alloy, Proc. Eng. 81 (2014) 610–615. [6] G. Bajargan, G. Singh, D. Sivakumar, U. Ramamurty, Effect of temperature and strain rate on the deformation behavior and microstructure of a homogenized AZ31 magnesium alloy, Mater. Sci. Eng. A 579 (2013) 26–34. [7] A. Gledhill, Correlation of plastic deformation and dynamic recrystallization in magnesium alloy ZK60, Acta Mater. 49 (2001) 1199–1207. [8] M.R. Barnett, Z. Keshavarz, M.D. Nave, Microstructural features of rolled Mg-3Al1Zn, Metall. Mater. Trans. A 36 (2005) 1697–1704. [9] M.R. Barnett, N. Stanford, Influence of microstructure on strain distribution in Mg–3Al–1Zn, Scr. Mater. 57 (2007) 1125–1128. [10] X. Huang, K. Suzuki, Y. Chino, M. Mabuchi, Influence of initial texture on rolling and annealing textures of Mg–3Al–1Zn alloy sheets processed by high temperature rolling, J. Alloys Compd. 537 (2012) 80–86. [11] M. Mabuchi, Y. Chino, H. Iwasaki, T. Aizawa, K. Higashi, The grain size and texture dependence of tensile properties in extruded mg-9Al-lZn, Mater. Trans. 42 (2001) 1182–1188. [12] C. Xu, M.Y. Zheng, K. Wu, E.D. Wang, G.H. Fan, S.W. Xu, S. Kamado, X.D. Liu, G.J. Wang, X.Y. Lv, Influence of rolling temperature on the microstructure and mechanical properties of Mg–Gd–Y–Zn–Zr alloy sheets, Mater. Sci. Eng. A 559 (2013) 615–622. [13] W.J. Kim, S.I. Hong, Y.S. Kim, S.H. Min, H.T. Jeong, J.D. Lee, Texture development and its effect on mechanical properties of an AZ61 Mg alloy fabricated by equal channel angular pressing, Acta Mater. 51 (2003) 3293–3307. [14] X. Huang, K. Suzuki, N. Saito, Textures and stretch formability of Mg–6Al–1Zn magnesium alloy sheets rolled at high temperatures up to 793 K, Scr. Mater. 60 (2009) 651–654. [15] S.E. Ion, F.J. Humphreys, S.H. White, Dynamic recrystallisation and the development of microstructure during the high temperature deformation of magnesium, Acta Metall. 30 (1982) 1909–1919. [16] J.M. Zhang, Z.H. Wang, B.L. Jiang, Z.S. Chen, L.B. Niu, Influence of annealing treatment on precipitation of Î2-Mg17Al12 phase of AZ80 magnesium alloy, Adv. Mater. Res. 476-478 (2012) 55–58. [17] P. Molnár, A. Jäger, P. Lejček, The role of low-angle grain boundaries in multi-
[22] [23] [24]
[25]
[26] [27] [28]
[29]
[30]
[31] [32] [33]
[34]
[35] [36] [37] [38]
48
temperature equal channel angular pressing of Mg–3Al–1Zn alloy, J. Mater. Sci. 47 (2011) 3265–3271. J. Koike, Enhanced deformation mechanisms by anisotropic plasticity in polycrystalline Mg alloys at room temperature, Metall. Mater. Trans. A 36 (2005) 1689–1696. S.W. Xu, S. Kamado, N. Matsumoto, T. Honma, Y. Kojima, Recrystallization mechanism of as-cast AZ91 magnesium alloy during hot compressive deformation, Mater. Sci. Eng. A 527 (2009) 52–60. M.R. Barnett, Z. Keshavarz, A.G. Beer, X. Ma, Non-Schmid behaviour during secondary twinning in a polycrystalline magnesium alloy, Acta Mater. 56 (2008) 5–15. J.L. Haughton, The technology of magnesium and its alloys, Nature 148 (1941) 67–68. H. Wang, P.D. Wu, M.A. Gharghouri, Effects of basal texture on mechanical behaviour of magnesium alloy AZ31B sheet, Mater. Sci. Eng. A 527 (2010) 3588–3594. Y.N. Wang, J.C. Huang, Texture analysis in hexagonal materials, Mater. Chem. Phys. 81 (2003) 11–26. S.W. Xu, K. Oh-ishi, S. Kamado, T. Homma, Twins, recrystallization and texture evolution of a Mg–5.99Zn–1.76Ca–0.35Mn (wt.%) alloy during indirect extrusion process, Scr. Mater. 65 (2011) 875–878. N. Stanford, M.R. Barnett, The origin of “Rare Earth” texture development in extruded Mg-based alloys and its effect on tensile ductility, Mater. Sci. Eng. A 496 (2008) 399–408. J.A.D. Valle, M.T. Pérez-Prado, O.A. Ruano, Texture evolution during large-strain hot rolling of the Mg AZ61 alloy, Mater. Sci. Eng. A 355 (2003) 68–78. M.H. Yoo, Slip, twinning, and fracture in hexagonal close-packed metals, Metall. Trans. A. 12 (1981) 409–418. Z. Wang, Z. Chen, C. Zhan, L. Kuang, J. Shao, R. Wang, C. Liu, Quasi-static and dynamic forced shear deformation behaviors of Ti-5Mo-5V-8Cr-3Al alloy, Mater. Sci. Eng. A 691 (2017) 51–59. J.L. Sun, P.W. Trimby, F.K. Yan, X.Z. Liao, N.R. Tao, J.T. Wang, Shear banding in commercial pure titanium deformed by dynamic compression, Acta Mater. 79 (2014) 47–58. H. Li, E. Hsu, J. Szpunar, H. Utsunomiya, T. Sakai, Deformation mechanism and texture and microstructure evolution during high-speed rolling of AZ31B Mg sheets, J. Mater. Sci. 43 (2008) 7148–7156. N.J. Petch, The cleavage strength of polycrystals, J. Iron Steel Inst. 174 (1953) 25–28. N. Hansen, Hall–Petch relation and boundary strengthening, Scr. Mater. 51 (2004) 801–806. S.M. He, X.Q. Zeng, L.M. Peng, X. Gao, J.F. Nie, W.J. Ding, Microstructure and strengthening mechanism of high strength Mg–10Gd–2Y–0.5Zr alloy, J. Alloys Compd. 427 (2007) 316–323. A.S. Khan, A. Pandey, T. Gnäupel-Herold, R.K. Mishra, Mechanical response and texture evolution of AZ31 alloy at large strains for different strain rates and temperatures, Int. J. Plast. 27 (2011) 688–706. X. Wu, Y. Zhu, Heterogeneous materials: a new class of materials with unprecedented mechanical properties, Math. Res. Lett. (2017) 1–6. I. Basu, T. Al-Samman, Twin recrystallization mechanisms in magnesium-rare earth alloys, Acta Mater. 96 (2015) 111–132. A.M. Galiyev, R.O. Kaibyshev, G. Gottstein, Grain Refinement of ZK60 Magnesium Alloy During Low Temperature Deformation, Magnesium Technol. (2002) 181–185. F. Guo, D. Zhang, X. Yang, L. Jiang, S. Chai, F. Pan, Influence of rolling speed on microstructure and mechanical properties of AZ31 Mg alloy rolled by large strain hot rolling, Mater. Sci. Eng. A 607 (2014) 383–389.