Journal of Materials Science & Technology, 2020, 52(0): 152-161 DOI: 10.1016/j.jmst.2020.04.022

Research Article

Investigation of Portevin-Le Chatelier effect in rolled α-phase Mg-Li alloy during tensile and compressive deformation

Xiaoqiang Lia, Chunlong Chenga, Qichi Le,a,*, Lei Baoa, Peipeng Jinb, Ping Wanga, Liang Rena, Hang Wanga, Xiong Zhoua, Chenglu Hua

Key Laboratory of Electromagnetic Processing of Materials, Ministry of Education, Northeastern University, Shenyang, 110819, China

Qinghai Provincial Key Laboratory of New Light Alloys, Qinghai Provincial Engineering Research Center of High Performance Light Metal Alloys and Forming, Qinghai University, Xining, 810016, China

Corresponding authors: *. E-mail address:qichil@mail.neu.edu.cn(Q. Le).

Received: 2019-12-19   Accepted: 2020-02-20   Online: 2020-09-15

Abstract

Avoiding the Portevin-Le Chatelier (PLC) effect is very important concern for wrought Mg-Li alloys. In this study, the special PLC effect was found in rolled Mg-5Li-3Al-2Zn (LAZ532) alloy during tensile and compressive deformation. By observing microstructure evolution of the alloy during tensile and compressive deformation, it was found that prismatic <a> and pyramidal <c + a> slips were activated during tensile deformation, resulting in plenty of dislocation accumulation. In the deformation process after compressive yielding, the deformations in coarse grains and fine grains were dominated by {10 $\bar{1}$ 2} extension twinning and grain boundary slip, respectively. Based on experimental result analysis, the sudden appearance of PLC effect in the later stage of axial tensile deformation (along rolled direction) was caused by interaction between solute atoms and dislocations. In the process of axial compressive deformation (along rolled direction), PLC effect presented the complex and changeable phenomenon of appeared-disappeared-appeared, which was mainly caused by the continuous nucleation of twin in the material, the activation of grain boundary slip and the shear deformation of twin, respectively.

Keywords: Mg-Li alloy ; Portevin-Le Chatelier effect ; Dynamic strain aging ; Twins

PDF (6710KB) Metadata Metrics Related articles Export EndNote| Ris| Bibtex  Favorite

Cite this article

Xiaoqiang Li, Chunlong Cheng, Qichi Le, Lei Bao, Peipeng Jin, Ping Wang, Liang Ren, Hang Wang, Xiong Zhou, Chenglu Hu. Investigation of Portevin-Le Chatelier effect in rolled α-phase Mg-Li alloy during tensile and compressive deformation. Journal of Materials Science & Technology[J], 2020, 52(0): 152-161 DOI:10.1016/j.jmst.2020.04.022

1. Introduction

With the increasing call for energy conservation and emission reduction, light metal alloys have attracted extensive attention of researchers in recent years [[1], [2], [3], [4]]. As the lightest structural metal material, magnesium-lithium (Mg-Li) alloys have been widely used in aerospace, automobile, electronics and other industrial fields [[5], [6], [7]]. As for Mg-Li alloys, addition of Li not only can effectively reduce the density of Mg alloys, but also can decrease the c/a ratio of Mg lattice to activate more slip systems [8,9]. This makes it have a series of advantages such as excellent processing formability and lower density than other Mg alloys.

At present, the addition of alloying elements is one of the main research hotspots of Mg-Li alloys. Al element was often used as a precipitation and solid solution hardener in Mg-Li alloys [10,11], and Zn was added to improve cold processing ability [12]. Moreover, Chen et al. [13] and Guo et al. [14] reported that comprehensive mechanical properties of Mg-Li alloys could be also improved by additions of a small amount of rare earth elements. For instance, the Mg-5Li-3Al-1.5Zn-2RE alloys fabricated by Li et al. [15] showed outstanding mechanical properties and fatigue properties, and tensile strength and fatigue strength were up to 282 MPa and 130 MPa (at 107 cycles), respectively. The α, α + β and β structures containing Mg-xLi-3Al-2Zn-0.2Y (x = 5,8,11) alloys prepared by Tang et al. [16] exhibited high strength and good plasticity, and tensile strength and tensile fracture strain were 310 MPa and 26%, respectively. The above studies showed that although the addition of alloying elements could improve some properties of Mg-Li alloys, the PLC effect was observed in the tensile stress-strain curve [15,17], which is very unfavourable for the application of the Mg-Li alloys because it is easy to cause premature failure of alloys. In previous studies, Wu et al. [18] found that the PLC effect in the β-phase Mg-14.3Li-0.8Zn alloy occurred throughout the stretch deformation, and the dynamic strain aging (DSA) mechanism was used to explain why the PLC effect occurred in solid-solution treated Mg-14.3Li-0.8Zn alloys bearing load under appropriate strain rate and temperatures. Furthermore, Li et al. [19] studied the PLC effect of Mg-5Li-3Al-1.5Zn-2RE alloy in the process of tensile deformation, and found that the traditional DSA was main cause of small serrated flow, while the severe PLC effect was attributed to a large number of twin nucleation. However, in the Mg-5Li-3Al-2Zn-0.2Y alloy prepared by Tang et al. [17], the sudden PLC effect occurred in the late stage of tensile deformation. This can lead to the sudden failure of the alloys, and is not conducive to timely preventive measures in the application process. In the above studies, there are no in-depth study on the sudden PLC effect. In addition, the current studies of PLC effect were focused on the process of tensile deformation, while there are few studies on the PLC effect in compression deformation process. Therefore, in order to effectively prevent the occurrence of various PLC effect in the process of tension and compression deformation, the causes of PLC effect occurrence need to be further studied.

In this study, the sudden and intermittent PLC effects of Mg-5Li-3Al-2Zn alloy were studied during tensile and compressive deformation at room temperature, respectively.

2. Experiment method

In this study, the experimental materials were commercial pure Mg ingots (99.9%), pure Li strips (99.9%), pure zinc ingots (99.9%) and Mg-Al alloy powder (50 wt% Mg, 50 wt% Al, 100-150 mesh, Tangshan Weihao Magnesium Powder Co., Ltd). The extruded Mg-5Li-3Al-2Zn (LAZ532) alloy (ø12 mm) was prepared by casting followed by hot-extrusion at 573 K, and the process was described in literature specifically [20]. Then, the extruded square bars were further rolled, and the process was as follows: firstly, the extruded squares rods were kept for 0.5 h in the heating furnace at 573 K, then rolled at different passes (3, 6, and 9 passes) at 573 K. Finally, the rolled plates with a thickness of 3 mm were obtained. The rolled plates were kept warm for 5-10 min in heating furnace at 573 K, and then quenched by water to release internal stress.

Instron-5982 testing machine was used to measure mechanical properties of the rolled LAZ532 alloy, and both stretching and compressive rates were 0.2 mm min-1. Both tensile and compressive loads were applied along the rolled direction. Electron back-scattered diffraction (EBSD) was performed by scanning electron microscopy (SEM, ZEISS-6035 field-emission) to analyse microstructure of the alloy. The EBSD data were analysed by using an Oxford HKL Channel 5 software. The detailed preparation process of the EBSD samples was as follows: Firstly, the samples were mechanically polished, followed by electrochemical polishing with 7 vol.% perchloric acid alcohol solution to eliminate internal stress on the surface of samples (electrochemical polishing parameters: temperature is -30 ℃, voltage is 20 V, electric current is 0.2 mA, time is 2 min); then argon-ion polishing was performed to further eliminate residual internal stress and perchloric acid alcohol solution on the surface of samples (argon-ion polishing parameters: voltage is 4 kV, ion beam inclination is 4°, time is 90 min). In addition, microstructure evolution of the alloy during compression deformation was observed by in-situ SEM. The preparation process of in-situ SEM samples was as follows: firstly, a cuboid block of 4 mm × 3 mm × 3 mm (length × width × thickness) was taken along the rolling direction by wire cutting, then the rolling surface was mechanically polished until bright and traceless, and etched with a solution consisting of 10 mL acetic acid, 4.2 g picric acid, 70 mL alcohol and 10 mL distilled water. Finally, the dislocation in the stretched sample was observed by transmission electron microscopy (TEM, JEM-2100 F). The preparation process of TEM sample was as follows: firstly, 1 mm thick disc was cut by wire cutting and grinded to a thickness of 80 μm with sandpaper; then the disc was punched into slice of 3 mm by punching machine, and then the TEM samples with thin area were prepared by electrolysis double spray (electrolytic double spray parameters: 4 vol.% perchloric acid alcohol, current is 4 mA); Finally, the TEM samples were refined by ion shear thinning (ion shear thinning parameters: voltage is 1Kv, ion beam inclination is 1°, time is 20-30 min).

3. Results and discussion

3.1. Mechanical properties

Fig. 1 shows the stress-strain curves of LAZ532 alloys at different rolling passes. As shown in the local magnified images of Fig. 1(a) and (b), the PLC effect, or the so-called “serrated flow” appeared in all tensile and compressive curves of the alloys. This phenomenon is a continuous process of strain localization; the stress-strain curve of the alloy shows plastic instability or serrated fluctuation [21]. In macroscopic view, it results in the reduction of plasticity and convex-concave surface of deformed part [22]. Therefore, this plastic instability should be avoided as far as possible in structural alloys. As shown in the local magnified image of Fig. 1(a), when the tensile stress reaches the maximum, an apparently stable serrated stress flow pops up, the serration type is identified as type B, in which the serrations oscillate about the general level of the stress-strain curve [23]. As shown in local magnified image of Fig. 1(b), when the compressive strain reaches yielding point, a slight increasing stress fluctuation appears, which is identified as type A, in which the serrations rise above the general level of the stress-strain curve [23]. However, when the compressive strain increases to about 8%, the serrated flow disappears, and the serrated flow appears again near the strain of 13% and then until compression fracture.

Fig. 1.

Fig. 1.   Mechanical properties of LAZ532 alloys at different rolling passes: (a) tensile properties; (b) compressive properties.


To further analyze the PLC effect, statistical distribution of stress fluctuations range for the alloys at different rolling passes is shown in Fig. 2. The stress fluctuations can be represented by the absolute value of the stress difference between adjacent the wave crest and the wave trough (Δσ), which can be calculated by Δσ=σcrest-σtrough. Then, make frequency statistics for all calculated Δσ values with a class interval of 0.5 MPa. As can be seen from Fig. 2(a), the distributions of amplitudes of stress fluctuations in the range of tensile strain 16%-18% are bell-shaped histogram, and the values of Δσ gather in the range of 3-4 MPa. Fig. 2(b) presents that the results of a low-amplitude oscillations were observed in the range of compressive strain 16%-20%, and all plots display a high probability of very small stress fluctuations near the origin. The low-amplitude oscillations cannot be attributed to some kind of random noise because they are described by a power law [24,25], which proves a non-stochastic nature of low-amplitude stress fluctuations as described above. In addition, the PLC effect is related to dynamic strain ageing (DSA) mechanism and shear deformation of twins [19]. However, in this study, the sudden and intermittent PLC effects in the process of tension and compression deformation are quite different from the PLC effect found by Li et al. [19] and Wu et al. [18] in Mg-Li alloys, and there is no relevant literature report at present.

Fig. 2.

Fig. 2.   Distribution of amplitudes of stress fluctuations for LAZ532 alloys at different rolling passes: (a) the range from 16% to 18% tensile strain; (b) the range from 16% to 20% compressive strain.


In addition, to sum up, it can be seen from the analysis results in Fig. 1, Fig. 2 that the PLC effect appeared in LAZ532 alloys with different rolling passes, and the type and characteristics of the PLC effect are the same. This indicates that the occurrence of the PLC effect is not accidental, and the rolling pass has little effect on PLC effect. Therefore, in order to further investigate the causes of the PLC effect, the LAZ532 alloy rolled 9 times as a representative for a detailed analysis in the latter part.

3.2. Microstructure analysis

Fig. 3, Fig. 4 show the EBSD images, micro-texture evolution and twin types maps of LAZ532 alloy after rolled 9 passes during tensile and compressive deformation. RD, TD and ND in Fig. 3 are the rolling direction, transverse direction and normal direction, respectively. The red and blue grain boundaries in Fig. 4 represent {10$\bar{1}$ 2} extension twinning and {10$\bar{1}$ 1} contraction twinning, respectively. As shown in Fig. 3(a) and (b), the grain orientation of LAZ532 alloy rolled 9 passes is mainly in the direction of <0001>, that is, <0001> of grains is parallel to ND. Compared to LAZ532 alloy rolled 9 passes with 0% tensile strain, as shown in Fig. 3(c) and (d), the <0001> of grains is still parallel to ND and the grain orientation shows no significant change when tensile strain is 6%. The texture intensity values increase from 7.28 to 8.49. Besides, as shown in Fig. 4(a) and (b), there were no obvious twins inside the grains. As tension strain further increases to 17%, as shown in Fig. 3(e) and (f), the texture intensity value increases to 10.54. This indicates that the concentration of (0001) plane perpendicular to ND increases during tensile deformation, but texture type is still {0001} basal lamellar texture. Furthermore, as shown in Fig. 4(c) and (d), although twins were not observed, a large number of low angle grain boundaries (LAGBs) appear in grain interior. This indicates that there are plenty of dislocation in the alloy because the grain boundaries prone to tilting and twisting in the deformation process to form sub-grain boundary [26]. As shown in Fig. 3(g) and (h), when compressive strain is 6%, abundant twins were observed in coarse grains, and the texture intensity value decreases to 6.36 due to twins formation and fine grains rotation [20]. According to geometric condition, {10$\bar{1}$ 2} extension twins in Mg alloys easily occurs when (0001) plane of grains is subjected to tensile stress, and there is an 86° between the twin and their parent [27]. As shown in Fig. 4(e) and (f), the misorientation angle is clustered around 86°. Therefore, for the rolled LAZ532 alloy with {0001} basal lamellar texture, {10 $\bar{1}$ 2} extension twinning only occurs during the compressive deformation along the RD.

Fig. 3.

Fig. 3.   EBSD images and inverse pole figures of LAZ532 alloy rolled 9 passes during tensile and compressive deformation: (a, b) initial state; (c, d) 6% tensile strain; (e, f) 17% tensile strain; (g, h) 6% compressive strain.


Fig. 4.

Fig. 4.   Twins and misorientation distribution maps of LAZ532 alloy rolled 9 passes during tensile and compressive deformation: (a, b) 6% tensile strain; (c, d) 17% tensile strain; (e, f) 6% compressive strain.


In order to analyze the orientation of fine grains and twins, as shown in Fig. 5(a) and (b), the orientation of fine grains and twins change significantly. The weak pole density points of (0001) pole figure are distributed along two sides of RD, and EBSD maps of corresponding pole density points (black circles) are shown in Fig. 5(c) and (d). As can be seen from Fig. 5(c) and (d), the number ratio of between twins and fine grains is close to 3:1. This indicates that the deformation at this time is dominated by twinning and accompanied by the rotation of (0001) plane of some fine grains in the direction perpendicular to RD. It is consistent with our previous results [20]. In addition, in order to further visually observed microstructure evolution during compression deformation. Fig. 6 shows the SEM maps of LAZ532 alloy rolled 9 passes at in-situ compression deformation process. As shown in Fig. 6(a), at the initial stage (0% compressive strain), coarse and fine grains coexist in Mg matrix, and there is no obvious distortion in grains. As the increasing of compression strain, as shown in Fig. 6(b)-(d), the interior of coarse grains in two representative regions (Z1 and Z2) became unsmooth and twins were obviously observed, meanwhile, fine equiaxed grains (G1, G2, G3 and G4) gradually evolve into approximately ellipse, and generate obvious deflection (about the range of 8°-10°). This further proves that the deformation in coarse grains and fine grains are dominant by twin nucleation and grain boundary slip, respectively, in the deformation process after compressive yielding. Besides, it can be seen from Fig. 6(a)-(d) that in the process of compression deformation, a handful of white precipitated phase in matrix have little effect on the nucleation of twins in coarse grains and the rotation of fine grains. Therefore, the above results show that the nucleation of twins and the rotation of fine grains during the compression deformation may be related to the intermittent PLC effect (Fig. 1(b)), and the effect of precipitation on PLC effect can be ignored because of its the extremely low content.

Fig. 5.

Fig. 5.   (a) EBSD image and (b) pole figures of fine grains and twinning areas of LAZ532 alloy rolled 9 passes at 6% compressive strain, and (c, d) EBSD maps of corresponding pole density points (black circles).


Fig. 6.

Fig. 6.   In-situ SEM maps of LAZ532 alloy rolled 9 passes during compressive deformation: (a) initial state; (b) 5% strain; (c) 10% strain; (d) 13% strain.


Local misorientation could be measured by EBSD to study the geometrically necessary dislocations (GND) [28,29]. Fig. 7 shows that the local misorientation maps and local misorientation angle distribution maps of LAZ532 alloy after rolled 9 passes during tensile and compressive deformation. Fig. 7(a) and (b) shows the local misorientation of LAZ532 alloy rolled 9 passes (initial state), and local misorientation angle calculated less than 5° was included in this study. Compared to the LAZ532 alloy rolled 9 passes (initial state), as shown in Fig. 7(c)-(f), with the increasing of tensile strain, the green color in local misorientation maps gradually deepen, and the local misorientation angles obviously increase. In addition, as shown in Fig. 7(g) and (h), the local misorientation also significantly increases in compression deformation process. The local misorientation was used to calculate the GND values by a simple method equation [30,31]:

${{\rho }^{\text{GND}}}=\frac{2\Delta {{\theta }_{i}}}{ub}$

where ρGND is the GND density at the point of interest; Δθi represents the local misorientation; u is the unit length of the point (400 nm); b is the Burgers vector (for Mg, it is 3.21 × 10-10 m). The column charts of calculated GND density were given in Fig. 8. The results indicate that the GND inside the material gradually increases when level of plastic deformation increases (Fig. 8(b)-(d)).

Fig. 7.

Fig. 7.   Local misorientation maps and local misorientation angle distribution maps of LAZ532 alloy rolled 9 passes during tensile and compressive deformation: (a, b) initial state; (c, d) 6% tensile strain; (e, f) 17% tensile strain; (g, h) 6% compressive strain.


Fig. 8.

Fig. 8.   Calculated mean GND density distributions of LAZ532 alloy rolled 9 passes during tensile and compressive deformation: (a) initial state; (b) 6% tensile strain; (c) 17% tensile strain; (d) 6% compressive strain.


Relevant studies have shown that dense dislocation in the deformation process is easy to be pinned by solute atoms, resulting in serrate oscillation of the stress-strain curve [18,19,32]. However, the formation of dense dislocation in Mg alloy need to activate prismatic <a> and pyramidal <c+a> slips, and the opening of each slip system is directly related to SF [33]. Fig. 9 shows the SF distribution maps of Mg-5Li-3Al-2Zn alloy rolled 9 passes during tensile deformation. As shown in Fig. 9(a) and (d), at the initial stage, the average SF of prismatic <a> and pyramidal <c+a> slips (0.44) is higher than that of basal <a> slip (0.21). With the increasing of tensile strain, as can be seen from Fig. 9(b)-(d) the average SF of pyramidal slip gradually increases to 0.47, and the average SF of prismatic slip basically remains unchanged. Meanwhile, the SF of basal slip slightly decreases. This indicates that with the increase of tensile strain, the deformation is dominated gradually by pyramidal slip, and next prismatic slip because Mg-Li alloys have a low axial ratio (c/a) [8,9]. The activation of prismatic <a> and pyramidal <c+a> slips can prevent the formation of local dislocation by basal <a> dislocation decomposition because of Mg alloys with low basal stacking fault energy [34]. It is favorable for dislocation climbing and cross slip to form high density dislocation accumulation, which is easily pinned by solute atoms, resulting in the occurrence of PLC effect. Therefore, in this study, the high density dislocation formed in the late stage of tensile deformation, which is associated with sudden PLC effect (Fig. 1(a)).

Fig. 9.

Fig. 9.   Schmid factor (SF) distribution maps of Mg-5Li-3Al-2Zn alloy rolled 9 passes during tensile deformation: (a) initial state; (b) 6% tensile strain; (c) 17% tensile strain; (d) average SF values of each slip system under different strain.


In order to observe dislocation accumulation of the alloy after tensile deformation. TEM maps of LAZ532 alloy rolled 9 passes at 17% tensile strain are shown in Fig. 10. According to g∙b=0 invisibility criterion, two-beam dark-field images of the deformed microstructure were obtained near [0001] and [1$\bar{2}$ 1 $\bar{3}$] zones axis under g = (1 $\bar{2}$ $\bar{1}$ 0), g = (01 1-0) and g = (0 1-11), g = (10 $\bar{1}$ 0), respectively. Although both <a> and <c+a> dislocations are visible under the above conditions [33], as can be seen from Fig. 10(a)-(d), a large number of twisted dislocations were observed to be entangled, which is prone to form <c+a> dislocations. Subsequently, <c+a> dislocations most likely are blocked by faults causing stacking fault formation. Consequently, this results plenty of forest dislocations are formed.

Fig. 10.

Fig. 10.   TEM maps of LAZ532 alloy rolled 9 passes at 17% tensile strain: (a, b) two-beam dark-field images taken along the [0001] zone axis under g = (1- 2 $\bar{1}$ 0) and g = (01 $\bar{1}$ 0), respectively; (c, d) two-beam dark-field images taken along the [1 $\bar{2}$1 $\bar{1}$] zone axis under g = (0 $\bar{1}$ 11) and g = (10 $\bar{1}$ 0), respectively.


4. Discussion

The phenomenon of serrated flow is noteworthy in this study. As shown in Fig. 1(a), the sudden PLC effect was observed in all of rolled LAZ532 alloys during the tensile deformation. The DSA mechanism can explain the occurrence of the PLC effect during tensile deformation [18,32], which is the interaction between mobile dislocations and solute atoms in the process of slip deformation. In order to further study its mechanism, schematic diagrams of solute atoms, {0001}<11 $\bar{2}$ 0> slip system, schematic illustration of deformation mechanisms and the DSA model are shown in Fig. 11. As shown in Fig. 11(a) and (b), Li, Al and Zn atoms are dissolved in the (11$\bar{2}$0), (10 $\bar{1}$0), and (11 $\bar{2}$2) planes of Mg lattice [35,36], respectively, and the < a > slip direction of Mg alloy is <11$\bar{2}$0 > . As shown in Fig. 11(c), when stretching along the RD, the (0001) plane of grains rotated a direction perpendicular to ND because of the existence of {0001} basal lamellar texture (Fig. 3(b)), and the deformation is mainly prismatic <a> and pyramidal <c+a> slips (Fig. 9), that is, the slip directions are <11 $\bar{2}$0> and <11 $\bar{2}$3 >, thus forming a plenty of forest dislocation [37]. This increases the probability of dislocation pinning by solute atoms during the slip deformation, and also explains why the PLC effect occurs easily in Mg-Li alloys. Moreover, Gilman et al. [38] and Lloyd [39] et al. deemed that the movement of dislocation is discontinuous during deformation. When moving ability of mobile dislocation matches moving ability of solute atoms, the mobile dislocation will stay for a short time due to the presence of obstacles [40]. Therefore, as shown in Fig. 11(d), at the initial stage of tensile deformation, solute atoms diffuse slowly and dislocation density is lower (Figs. 7(c) and (d) and 8 (b)), which is not enough to pin dislocation. At this time, it is not enough to produce PLC effect, so the initial curve of tensile deformation is smooth. With the increasing of tensile strain, the vacancy concentration in the material increases sharply, which accelerates the diffusion rate of solute atoms; Solute atoms near mobile dislocation migrate to mobile dislocation by means of pipeline diffusion, resulting in enrichment of solute atoms near mobile dislocation [41]. Meanwhile, dislocation density increases sharply (Figs. 7(e) and (f), 8 (c) and 10) causes dislocation to be locked by solute atoms. At this point, if dislocation continues to move, a greater external force is needed to get rid of the pinning of solute atoms, and then dislocation is pinned by solute atoms again, thus go round and begin again, causing the stress-strain curve become serrated.

Fig. 11.

Fig. 11.   (a) Schematic diagram of solute atoms, (b) {0001}<11 $\bar{2}$ 0> slip system, (c) schematic illustration of the grain rotation mechanisms during deformation and (d) the dynamic strain aging model.


As can be seen from Fig. 1(b), the intermittent PLC effect was also observed during compressive deformation. Furthermore, a large number of {10 $\bar{1}$ 2} extension twins were found in coarse grains of LAZ532 alloy with 6% compression strain (Figs. 3(g) and 4 (e)), and dislocation density of the sample with 6% compression strain is slightly lower than that of the sample with 6% tension strain (Fig. 8(b) and (d)). There is no “serrated flow” phenomenon in the tension curve at 6% strain, so dislocation density in the 6% compression sample is not enough to cause the PLC effect. Relevant studies shown that the cause of the PLC effect occurrence is not only the interaction between solid solution atoms and dislocation, but also the instantaneous strain provided by shear deformation of twins [18,19,32]. In addition, the study of Lebedkina et al. [42] also showed that the PLC effect was easy to occur in Al 3Mg alloy with coarse grains because twins tend to nucleate in coarse grains during deformation. In this study, the deformation after compression yielding (2%-8% strain) is dominated by nucleation and expansion of twins, and twinning mechanism was shown in Fig. 11(c). It can be concluded that the PLC effect has important links with twins. In general cases, the stress required for twin nucleation is greater than that required for expansion, so the stress drops sharply after twin nucleation [43]. Therefore, stress-strain curve presents serrated due to the continuous formation of twins during deformation. With the increasing of strain (8%-13%), the content of twins in coarse grains reaches saturation, and fine grains begin to rotate (Fig. 5, Fig. 6) because {10 $\bar{2}$0} extension twinning is inhibited in fine grains [20,44]. At this time, the main deformation is dominated by grain boundary slip. It is difficult for solid solution atoms to pin dislocation at grain boundaries, so there was no serrated flow phenomenon in the stress-strain curve at this stage. As the strain increases to 13%, the (0001) plane of most of fine grains rotate to the direction perpendicular to RD (Fig. 11(c)). At this point, {10 $$\bar{1}$2} extension twinning and basal slip are inhibited and difficult to be activated, and the critical resolved shear stress (CRSS) of {10$\bar{1}$1} contraction twinning and (10 $\bar{1}$0) prismatic slips are also very high [45,46]. Therefore, with the strain continues to increase, slip deformation is difficult to continue, and then the stress begins to pass through twins. When the stress begins to pass through twins, the external applied stress increases, and then it decreases after the stress passes through twins. Thus go round and begin again until the compression fracture. This causes the stress-strain curve to become serrated.

5. Conclusion

Based on the experimental results of this work, in the rolled LAZ532 alloys, the sudden PLC effect during the tensile deformation was caused by the interaction between solid solution atoms and dense dislocation. However, the intermittent PLC effect during the compression deformation was caused by twin’s nucleation and the instantaneous strain provided by shear deformation of twins, and the main reason for the disappearance of the PLC effect was that the grain boundary slip deformation replaced the nucleation of twins. In addition, the PLC effect is also affected by deformation temperatures and rates. However, in Mg-Li alloys, the related studies are rarely reported, and the related mechanisms remain to be further studied.

Acknowledgments

The work was financially supported by the National Key Research and Development Program of China (No. 2016YFB0301104) and the National Natural Science Foundation of China (No. 51771043).

Reference

B.J. Wang, D.K. Xu, S.D. Wang, E.H. Han, Front. Mech. Eng., 14(2019), pp. 113-127.

DOI      URL     [Cited within: 1]

M.K. Kulekci, Int. J. Adv. Manuf. Technol., 39(2008), pp. 851-865.

DOI      URL     [Cited within: 1]

G.D. Barati, M. Aliofkhazraei, P. Hamghalam, N. Valizade, J. Magn. Alloy, 5(2017), pp. 74-132.

DOI      URL     [Cited within: 1]

B.J. Wang, D.K. Xu, S.D. Wang, L.Y. Sheng, R.C. Zeng, E.H. Han, Int. J. Fatigue, 120(2019), pp. 46-55.

DOI      URL     [Cited within: 1]

J. Zhao, Z.Q. Li, W.C. Liu, J. Zhang, L. Zhang, Y. Tian, G.H. Wu, Mater. Sci. Eng. A, 669(2016), pp. 87-94.

DOI      URL     [Cited within: 1]

S. Feng, W.C. Liu, J. Zhao, G.H. Wu, H.H. Zhang, W.J. Ding, Mater. Sci. Eng. A, 692(2017), pp. 9-16.

DOI      URL     [Cited within: 1]

B.J. Wang, J.Y. Luan, D.K. Xu, J. Sun, C.Q. Li, E.H. Han, Acta Metall. Sin. (Engl. Lett.), 32(2019), pp. 1-9.

DOI      URL     [Cited within: 1]

R.H. Li, F.S. Pan, B. Jiang, H.W. Dong, Q.S. Yang, Mater. Sci. Eng. A, 562(2014), pp. 33-38.

DOI      URL     [Cited within: 2]

H. Ji, X. Peng, X.L. Zhang, W.C. Liu, G.H. Wu, L. Zhang, W.J. Ding, J. Alloys Compd., 791(2019), pp. 655-664.

DOI      URL     [Cited within: 2]

F.W. Bach, M. Schaper, C. Jaschik, Mater. Sci. Forum,419-422(2003), pp. 1037-1042.

[Cited within: 1]

T.L. Zhang, T. Tokunaga, M. Ohno, M.L. Zhang, Acta Metall. Sin.(Engl. Lett.), 32(2019), pp. 169-177.

[Cited within: 1]

H. Takuda, S. Kikuchi, T. Tsukad, K. Kubota, N. Hatta, Mater. Sci. Eng. A, 271(1999), pp. 251-256.

DOI      URL     [Cited within: 1]

Z.Y. Chen, Z.Q. Li, C. Yu, Mater. Sci. Eng. A, 528(2011), pp. 961-966.

DOI      URL     [Cited within: 1]

J. Guo, L.L. Chang, Y.R. Zhao, Y.P. Jin, Mater. Charact., 148(2019), pp. 35-42.

DOI      URL     [Cited within: 1]

T.Q. Li, Y.B. Liu, Z.Y. Cao, D.M. Jiang, L.R. Cheng, Mater. Sci. Eng. A, 527(2010), pp. 7808-7811.

DOI      URL     [Cited within: 2]

Y. Tang, W.T. Jia, X. Liu, Q.C. Le, Y.L. Zhang, Mater. Sci. Eng. A, 675(2016), pp. 55-64.

DOI      URL     [Cited within: 1]

Y. Tang, Q.C. Le, W.T. Jia, L. Fu, X. Liu, J.Z. Cui, Mater. Sci. Eng. A, 704(2017), pp. 344-359.

DOI      URL     [Cited within: 2]

S.K. Wu, C. Chen, C.S. Yang, H.Y. Bor, Mater. Sci. Eng. A, 605(2014), pp. 33-38.

DOI      URL     [Cited within: 5]

T.Q. Li, Y.B. Liu, Z.Y. Cao, R.Z. Wu, M.L. Zhang, L.R. Cheng, D.M. Jiang, J. Alloys Compd., 509(2011), pp. 7607-7610.

DOI      URL     [Cited within: 5]

X.Q. Li, C.L. Cheng, Q.L. Le, X. Zhou, Q.Y. Liao, X.R. Chen, Y.H. Jia, P. Wang, J. Alloys Compd., 805(2019), pp. 947-956.

DOI      URL     [Cited within: 4]

D. Yuzbekova, A. Mogucheva, D. Zhemchuzhnikova, T. Lebedkina, M. Lebyodkin, R. Kaibyshe, Int. J. Plast., 96(2017), pp. 210-226.

DOI      URL     [Cited within: 1]

J. Zdunek, W.L. Spychalski, J. Mizera, K.J. Kurzydowski, Mater. Charact., 58(2007), pp. 46-50.

DOI      URL     [Cited within: 1]

H.F. Jiang, Q.C. Zhang, X.D. Chen, Z.J. Chen, Z.Y. Jiang, X.P. Wu, J.H. Fan, Acta Mater., 55(2007), pp. 2219-2228.

DOI      URL     [Cited within: 2]

A.H. CottrellDislocations and Plastic Flow in Crystals, Oxford University Press, Oxford(1953).

[Cited within: 1]

Y.L. Cai, S.L. Yang, S.H. Fu,Di Zhang,Q.C. Zhang, J. Mater. Sci. Technol., 33(2017), pp. 580-586.

DOI      URL     [Cited within: 1]

S. Fensin, D. Spearot, T. Bieler, S. Dillon, J. Luo, JOM, 71(2019), pp. 1198-1199.

DOI      URL     [Cited within: 1]

R.L. Xin, C.F. Guo, Z.R. Xu, G.D. Liu, X.X. Huang, Q. Liu, Scr. Mater., 74(2014), pp. 96-99.

DOI      URL     [Cited within: 1]

M. Calcagnottok, D. Ponge, E. Demir, D. Raabe, Mater. Sci. Eng. A, 527(2010), pp. 2738-2746.

DOI      URL     [Cited within: 1]

Z.F. Yan, D.H. Wang, X.L. He, W.X. Wang, H.X. Zhang, P. Dong, C.H. Li, Y.L. Li, J. Zhou, Z. Liu, L.Y. Sun, Mater. Sci. Eng. A, 723(2018), pp. 212-220.

DOI      URL     [Cited within: 1]

H. Gao, Y. Huang, W.D. Nix, J.W. Hutchinson, J. Mech. Phys. Solids, 47(1999), pp. 1239-1263.

DOI      URL     [Cited within: 1]

L.P. Kubin, A. Mortensen, Scr. Mater., 48(2003), pp. 119-125.

DOI      URL     [Cited within: 1]

S.M. Zhu, J.F. Nie, Scr. Mater., 50(2004), pp. 51-55.

DOI      URL     [Cited within: 3]

G.M. Zhu, L.Y. Wang, H. Zhou, J.H. Wang, Y. Shen, P. Tu, H. Zhu, W. Liu, P.P. Jin, X.Q. Zeng, Int. J. Plast., 120(2019), pp. 164-179.

DOI      URL     [Cited within: 2]

M.R. Barnett, Metall. Mater. Trans. A, 34(2003), pp. 1799-1806.

DOI      URL     [Cited within: 1]

C. Wang, H.Y. Zhang, H.Y. Wang, G.J. Liu, Q.C. Jiang, Scr. Mater., 69(2013), pp. 445-448.

DOI      URL     [Cited within: 1]

S.R. Agnew, M.H. Yoo, C.N. Tomé, Acta Mater., 49(2001), pp. 4277-4289.

DOI      URL     [Cited within: 1]

J.B. Lin, W.J. Ren, X.Y. Wang, Acta Metall. Sin., 52(2016), pp. 264-270, (in Chinese).

[Cited within: 1]

J.J. Gilman, J. Appl. Phys., 36(1965), pp. 3195-3206.

DOI      URL     [Cited within: 1]

D.J. Lloyd, P.J. Worthington, J.D. Embury, Philos. Mag., 21(1970), pp. 1147-1160.

DOI      URL     [Cited within: 1]

S. Bross,P. Hhner,E.A. Steck, Comput. Mater. Sci., 26(2003), pp. 46-55.

DOI      URL     [Cited within: 1]

C. Wang, Y.B. Xu, E.H. Han, Acta Metall. Sin., 42(2006), pp. 191-194, (in Chinese).

[Cited within: 1]

T.A. Lebedkina, M.A. Lebyodkin, T.T. Lamark, M. Janeek, Y. Estrin, Mater. Sci. Eng. A, 615(2014), pp. 7-13.

DOI      URL     [Cited within: 1]

A. Ostapovets, A. Serra, J. Mater. Sci., 52(2017), pp. 533-540.

DOI      URL     [Cited within: 1]

J.H. Wang, X.Q. Li, P.P. Jin, S.P. Li, G.J. Ma, L. Zhao, Mater. Res. Express, 5 (2018), Article 116518.

DOI      URL     PMID      [Cited within: 1]

H.Y. Wang, J. Rong, Z.Y. Yu, M. Zha, C. Wang, Z.Z. Yang, R.Y. Bu, Q.C. Jiang, Mater. Sci. Eng. A, 697(2017), pp. 149-157.

DOI      URL     [Cited within: 1]

A. Fernández, M.T.P. Prado, Y.J. Wei, A. Jérusalem, Int. J. Plast., 27(2011), pp. 1739-1757

DOI      URL     [Cited within: 1]

/