Journal of Materials Science & Technology  2019 , 35 (11): 2477-2484 https://doi.org/10.1016/j.jmst.2019.07.028

Orginal Article

Effects of icosahedral phase on mechanical anisotropy of as-extruded Mg-14Li (in wt%) based alloys

Chuanqiang Liabc, Daokui Xub*, Baojie Wangad**, Liyuan Shenga, Ruizhi Wue, Enhou Hanb*

aPeking University, Shenzhen Institute, Shenzhen Key Lab Human Tissue Regenerate & Repair, Shenzhen 518057, China
bKey Laboratory of Nuclear Materials and Safety Assessment, Institute of Metal Research, Chinese Academy of Sciences, Shenyang 110016, China
cSchool of Materials and Energy, Guangdong University of Technology, Guangzhou 510006, China
dSchool of Environmental and Chemical Engineering, Shenyang Ligong University, Shenyang 110159, China
eKey Laboratory of Superlight Materials & Surface Technology, Ministry of Education, Harbin Engineering University, Harbin 150001, China

Corresponding authors:   *Corresponding author.**Corresponding author at: Peking University, Shenzhen Institute, Shenzhen Key Lab Human Tissue Regenerate & Repair, Shenzhen 518057, China.E-mail addresses: dkxu@imr.ac.cn (D. Xu), bjwang@alum.imr.ac.cn (B. Wang),ehhan@imr.ac.cn (E. Han).*Corresponding author.**Corresponding author at: Peking University, Shenzhen Institute, Shenzhen Key Lab Human Tissue Regenerate & Repair, Shenzhen 518057, China.E-mail addresses: dkxu@imr.ac.cn (D. Xu), bjwang@alum.imr.ac.cn (B. Wang),ehhan@imr.ac.cn (E. Han).*Corresponding author.**Corresponding author at: Peking University, Shenzhen Institute, Shenzhen Key Lab Human Tissue Regenerate & Repair, Shenzhen 518057, China.E-mail addresses: dkxu@imr.ac.cn (D. Xu), bjwang@alum.imr.ac.cn (B. Wang),ehhan@imr.ac.cn (E. Han).

Received: 2019-02-25

Revised:  2019-04-8

Accepted:  2019-04-28

Online:  2019-11-05

Copyright:  2019 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

Through investigating and comparing microstructure and crystallographic texture of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y (in wt%) alloys, the differences in their mechanical anisotropy were investigated. It revealed that the formation of I-phase (Mg3Zn6Y, icosahedral structure) can effectively refine grain size. Moreover, compared with Mg-14Li alloy, the texture type of Mg-14Li-6Zn-1Y alloy changed slightly, but its texture intensity decreased remarkably. As a result, the stronger texture contributed to the “normal” mechanical anisotropy of Mg-14Li alloy with higher tensile strength and a lower elongation ratio along transverse direction (TD) than those along extrusion direction (ED). However, for Mg-14Li-6Zn-1Y alloy, the zonal distribution of I-phase particles along ED caused “abnormal” mechanical anisotropy, i.e. higher tensile strength and better plasticity along ED.

Keywords: Mg-Li alloys ; Texture ; I-phase ; Mechanical anisotropy ; Fracture

0

PDF (3050KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Chuanqiang Li, Daokui Xu, Baojie Wang, Liyuan Sheng, Ruizhi Wu, Enhou Han. Effects of icosahedral phase on mechanical anisotropy of as-extruded Mg-14Li (in wt%) based alloys[J]. Journal of Materials Science & Technology, 2019, 35(11): 2477-2484 https://doi.org/10.1016/j.jmst.2019.07.028

1. Introduction

Since density of lithium (Li) is 0.534 g/cm3, Mg-Li alloys are the lightest metallic materials with good plasticity, which ensures a great potential for the applications in a broad range of industries [[1], [2], [3], [4], [5]]. Generally, Mg-Li alloys have a typical body centered-cubic (BCC) structure and are composed of β-Li phases (i.e., Mg dissolution in Li matrix) when Li content is higher than 10.3 wt% [[6], [7], [8], [9], [10]]. Since BCC-structured Mg-Li alloys contain a relatively higher content of Li, their density can be further reduced [11]. Therefore, BCC-structured Mg-Li alloys have the highest specific strength [12].

However, Mg-Li alloys have low absolute strength and their ultimate tensile strength (UTS) is generally lower than 200 MPa even after being performed severe plastic deformation (SPD) [[13], [14], [15], [16]]. Recently, Xu et al. reported that when Zn and Y (with a Zn/Y atomic ratio of 6) were added, I-phase (Mg3Zn6Y) with an icosahedral quasicrystal structure could be in-situ formed in duplex structured Mg-Li-Zn-Y alloys [[17], [18], [19], [20]], which resulted in improved yield and ultimate tensile strengths to 166 MPa and 247 MPa [17]. Therefore, the I-phase strengthened Mg-Li alloys, as promising candidate materials, have a great potential in the engineering fields. Previous work demonstrated that for HCP-structured and (HCP + BCC) duplex structured wrought Mg-Li alloys, the formed strong texture can also cause their mechanical anisotropy [21,22]. For metallic materials, strong texture could cause the ridging phenomenon along rolling direction and earing effect during deep drawing process, which remarkably reduces their formability [23]. Lin et al. reported that due to plastic anisotropy in as-extruded ZK60 alloy sheet, drawn cups had obvious ears at the locations with an orientation angle of 45° with respect to the rolling direction [23]. For HCP-structured and (HCP + BCC) duplex structured Mg-Li-Zn-Y alloys, formation of I-phase could remarkably weaken the intensity of crystallographic texture and mechanical anisotropy [21,22]. Moreover, since plastic deformability of α-Mg phases is much lower than that of β-Li phases, previous work mainly focused on the effect of I-phase formation on texture evolution of α-Mg phases in both HCP-structured and duplex structured Mg-Li-Zn-Y alloys. For BCC-structured Mg-Li-Zn-Y alloys, matrix is only composed of β-Li phases. Then, effects of I-phase formation on texture evolution of β-Li phases and mechanical anisotropy should be considered. However, so far, no relevant literature can be referred.

Based on the description mentioned above, two questions can be proposed: (1) is it possible to form any I-phase in BCC-structured Mg-Li alloys? (2) if yes, how does it influence the crystallographic texture and mechanical anisotropy? This work aims to answer those questions through investigating and comparing microstructure and mechanical properties of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y (in wt%) alloys. Moreover, the underlying mechanisms for the differences in their mechanical anisotropy are discussed.

2. Experimental

The materials used herein are Mg-14Li and Mg-14Li-6Zn-1Y (in wt%) alloys prepared through a vacuum melting technique. After homogenization at 330 °C for 4 h, ingots with a diameter of 80 mm were extruded into plates with a cross-section of 50 mm (in width) × 15 mm (in thickness) at 300 °C. The extrusion ratio was about 6.7. Sample pieces cut from the extruded plates were ground with SiC papers up to 5000 grit finish and then finely polished up to a 1 μm finish with ethanol.

Under the condition of monochromatic CuKa radiation with a step size of 0.02° and a scan rate of data acquisition of 4°/min, main phases present in the alloys were analyzed by using a D/Max 2400 X-ray diffractometer (XRD). Microstructure of the alloys were observed by scanning electron microscopy (SEM; XL30-FEG-ESEM) and transmission electron microscopy (TEM; JEOL2100). TEM specimens were mechanically thinned and then performed argon ion milling. The polished surfaces were etched with an etchant consisting of 10% nitric acid and 90% ethanol and then observed by optical microscopy (OM) to characterize their grain structures. For texture analyses, samples with a rectangular cross-section of 20 mm (TD: transverse direction) × 25 mm (ED: extrusion direction) and a thickness of 5 mm were prepared. By using the Schultz reflection method of XRD analysis, the intensity contours of (110), (200) and (211) pole figures of two alloys were determined. Tensile samples with a gauge length of 20 mm and a cross-section of 3 mm × 2 mm were cut from the extruded plates. Tensile samples with their axial direction along the extrusion and transverse directions of the plates were designated as the “ED” and “TD” samples, respectively. Tensile experiments were conducted on a Care Measurement & Control testing machine at a constant strain rate of 1 × 10-3 s-1 at room temperature and all measurements were repeated at least three times. For tensile samples, one side of wide surfaces were finely polished before testing. Afterwards, the wide surfaces near to fractures and fracture characteristics were observed by SEM with the modes of secondary electron (SE) and backscattered electron (BSE), respectively.

3. Results and discussion

3.1. Microstructural characterization

XRD patterns of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y alloys are shown in Fig. 1. It can be seen that the Mg-14Li alloy is composed of β-Li phases. For Mg-14Li-6Zn-1Y alloy, besides β-Li phases, obvious peaks of formed I-phase can be detected. Fig. 2 shows SEM observations of polished surfaces of the alloys. It reveals that for Mg-14Li alloy, only β-Li matrix can be observed (Fig. 2(a) and (b)). For Mg-14Li-6Zn-1Y alloy, besides β-Li matrix, high densities of bright particles with the size of less than 5 μm are present and zonally distributed along ED direction (Fig. 2(c) and (d)). EDS results demonstrate that chemical compositions of these particles are mainly composed of Mg, Zn and Y (Fig. 2(e)). TEM observations demonstrate that the corresponding selected area diffraction patterns (SADP) of these particles exhibit typical characteristics of 5-fold symmetry, which further confirms the bright particles in Mg-14Li-6Zn-1Y alloy are I-phase particles (Fig. 2(f)).

Fig. 1.   XRD patterns of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y alloys.

Fig. 2.   Backscattered electron (BSE) images of (a) Mg-14Li and (c) Mg-14Li-6Zn-1Y alloys; (b) and (d) the high-magnification observations to squared areas in images (a) and (c), respectively; (e) EDS of the white particles in (d); (f) TEM observation to I-phase and its typical 5-fold diffraction pattern inserted.

Fig. 3 shows unveiled microstructure of the alloys through etching, revealing that Mg-14Li alloy consists of two different types of grain structures, i.e., (1) large elongated grains with length up to several millimeters (mm) and width of about 200 μm along ED, (2) small equiaxed grains with an average size of 90 μm distributed in interior of large elongated grains (Fig. 3(a)). Owing to the high chemical activity of Mg-14Li alloys, it is hard to etch out all grain boundaries (Fig. 3). For Mg-14Li-6Zn-1Y alloy, the average grain size is relatively fine (around 30 μm) and uniform.

Fig. 3.   Optical observations to the etched surface of (a) Mg-14Li and (b) Mg-14Li-6Zn-1Y alloys.

3.2. Texture analysis

Fig. 4 shows the {110}, {200} and {211} pole figures of the as-extruded Mg-14Li and Mg-14Li-6Zn-1Y alloys. It reveals that both alloys have a typical cube texture (i.e. {100}<001>). Meanwhile, the intensity of texture in Mg-14Li-6Zn-1Y alloy is 4.4 times lower than that of Mg-14Li alloy. In general, the texture intensity of Mg-Li alloys can be weakened when rare earth (RE) elements such as yttrium (Y) and neodymium (Nd) are added [24,25]. Moreover, the formed RE-containing phase particles can promote recrystallization and act as nucleation sites due to local inhomogeneity in strain energy and orientation caused by the particle-stimulated nucleation (PSN) effect [26]. Thus, the weaker texture of Mg-14Li-6Zn-1Y alloy can be mainly attributed to the formation of I-phase particles.

Fig. 4.   Pole figures of (a) Mg-14Li and (b) Mg-14Li-6Zn-1Y alloys.

3.3. Tensile properties and mechanical anisotropy

The typical engineering tensile stress-strain curves of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y alloys are shown in Fig. 5. To describe and compare the measured data conveniently, their mechanical properties in terms of 0.2% proof yield strength (YS), ultimate tensile strength (UTS) and elongation ratio to failure (EL) are summarized in Table 1. It can be seen that tensile strength of Mg-14Li alloy along TD is obviously higher than that along ED (Fig. 5(a)). YS and UTS of “TD” samples are 106 MPa and 131 MPa, whereas the associated values of “ED” samples are 82 MPa and 111 MPa, respectively. Meanwhile, EL of “TD” samples is only half of that of “ED” samples. In general, mechanical anisotropy is a reflection of high mechanical strength and low ductility in one direction, but they are reversed in another perpendicular direction [27]. However, for Mg-14Li-6Zn-1Y alloy, both the tensile strength and elongation of “TD” samples are much lower than those of “ED” samples (Fig. 5(b)). YS, UTS and EL of “TD” samples are respectively 137 MPa, 156 MPa and 13.5%, whereas the counterparts of “ED” samples are respectively 149 MPa, 168 MPa and 28.2%. This anomalous phenomenon was not observed in HCP-structured and (HCP + BCC) duplex structured Mg-Li based alloys [21,22]. Moreover, compared with Mg-14Li alloy, Mg-14Li-6Zn-1Y alloy has higher mechanical strength.

Fig. 5.   Engineering tensile stress-strain curves tested along the “ED” and “TD” directions of (a) Mg-14Li and (b) Mg-14Li-6Zn-1Y alloys.

Table 1   Mechanical properties of as-extruded Mg-14Li and Mg-14Li-6Zn-1Y alloys with different orientations.

AlloysOrientationYS (MPa)UTS (MPa)EL (%)
Mg-14LiED82 ± 3111 ± 540.8 ± 2 .4
TD106 ± 5131 ± 319.5 ± 1.8
Mg-14Li-6Zn-1YED149 ± 4168 ± 528.2 ± 2.1
TD137 ± 3156 ± 513.5 ± 1.5

New window

3.4. Failure analysis and fractography

The fracture surface morphologies of the alloys are shown in Fig. 6. It can be seen that for Mg-14Li alloy, the overall fracture surface of “ED” samples presents obvious necking characteristic (Fig. 6(a)) and obvious deep plastic dimples can be observed in the high-magnification image (Fig. 6(b)), whereas the necking degree of the overall fracture surface of “ED” samples reduces remarkably (Fig. 6(c)) and the size of plastic dimples is much larger (Fig. 6(d)). For as-extruded Mg-14Li-6Zn-1Y alloy, the overall fracture surfaces of both “TD” and “ED” samples are quite flat and necking features can hardly be observed (Fig. 6(e) and (g)). High-magnification images reveal that dimples on the fracture surfaces of “ED” samples are small and shallow, whereas numerous quasi-cleavage steps and no dimples are present on the fracture surfaces of “TD” samples (Fig. 6(f) and (h)). Thus, the cracking modes for “ED” and “TD” samples of Mg-14Li-6Zn-1Y alloy are respectively ductile and brittle fracture, resulting in higher EL of “ED” samples than that of “TD” samples (Fig. 5(b)). Since the large strain can be accommodated by deep plastic dimples during tensile tests [28], the elongation ratios of Mg-14Li alloy tested along both directions are much higher than those of Mg-14Li-6Zn-1Y alloy.

Fig. 6.   Observations to fracture surfaces of (a) and (c) “ED” and “TD” samples of Mg-14Li alloy, (e) and (g) “ED” and “TD” samples of Mg-14Li-6Zn-1Y alloy, (b), (d), (f) and (h) the high-magnification observations to the squared areas in images (a), (c), (e) and (g), respectively.

To further understand the deformation mechanisms occurred in “TD” and “ED” samples, the side surfaces close to the fractures were observed, as shown in Fig. 7. For Mg-14Li alloy, the presence of high densities of slip bands on the side surface of “ED” samples indicates that their deformation mechanism is mainly dominated by the dislocation slips (Fig. 7(a)). Meanwhile, the typical necking feature can be obviously observed. A high magnification image reveals that micro cracks preferentially nucleated along the slip bands (Fig. 7(b)). By contrast, the density of slip bands in the side surface of “TD” samples is lower and the necking feature can hardly be observed (Fig. 7(c) and (d)).

Fig. 7.   Observations to the side surfaces of (a) and (c) “ED” and “TD” samples of Mg-14Li alloy, (e) and (g) “ED” and “TD” samples of Mg-14Li-6Zn-1Y alloy. Images (b), (d), (f) and (h) the high-magnification observations to the squared areas in images (a), (c), (e) and (g), respectively.

For Mg-14Li-6Zn-1Y alloy, the slip bands were also present on the side surface of “ED” samples (Fig. 7(e)) and micro cracks mainly nucleated at slip bands (Fig. 7(f)), whereas the slip bands can hardly be observed on the surface of “TD” samples and cracks preferentially appeared at the zonally distributed I-phase particles (Fig. 7(g)). A high-magnification image reveals that the cracking mainly occurs at I-phase/β-Li interfaces (Fig. 7(h)).

3.5. Effect of I-phase formation on the grain structure

For Mg alloys, severe plastic deformation (SPD) was usually carried out at elevated temperatures, which can easily induce dynamic recrystallization (DRX) [29]. Previous work demonstrated that secondary phase particles in Mg alloys could provide more random nucleation sites for DRX grains and then result in a weakened texture [30,31]. Moreover, it has been reported that particles large than 1 μm in diameter can promote grain nucleation, whereas small-sized or close-spaced particles (with the interspace being less than 0.1 μm) will retard the nucleation of DRX grains [30,31]. In the current investigation, the size of most broken I-phase particles varies between 1 and 5 μm. Thus, those I-phase particles can induce the formation of stress concentration zones with a high density of dislocations and accelerate the recrystallization process of the surrounding matrix during hot extrusion, resulting in an obviously weaker texture intensity of Mg-14Li-6Zn-1Y alloy than that of Mg-14Li alloy (Fig. 4). Since Mg-14Li and Mg-14Li-6Zn-1Y alloys were subject to the same stress condition during the extrusion, the types of their texture were basically same (Fig. 4). Due to the presence of high stress/plastic strain fields, the nucleation of DRX grains is much easy at the interfaces between the matrix and I-phase particles. In addition, Bae et al. reported that microstructural coarsening and evolution can be effectively suppressed during the plastic deformation at elevated temperatures due to the presence of stable I-phase/matrix interfaces [32,33]. Thus, it can be concluded that the formation of a fine grain structure in Mg-14Li-6Zn-1Y alloy is mainly ascribed to the easy occurrence of recrystallization and the suppression to the growth of DRX grains induced by the existed I-phase particles.

3.6. Effect of I-phase formation on the mechanical strength and anisotropy

Based on the tensile results, it can be seen that YS of Mg-14Li-6Zn-1Y alloy tested along “ED” and “TD” are 1.82 and 1.29 times as high as those of Mg-14Li alloy, respectively. Generally, for metallic materials, the relationship between their yield strength and grain sizes is expressed by Hall-Petch equation [34].

σy = σ0 + kd-1/2 (1)

where d is the grain size, σ0 and k are the experimental constants. Since the two alloys contain different chemical compositions, they will have different constants of σ0 and k. Then, it is difficult to determine the concrete contribution of grain refinement to the improved strength of Mg-14Li-6Zn-1Y alloy. As for the I-phase, its effect on the strength improvement will be discussed as follows. Generally, I-phase is isotropic and possesses a specially ordered lattice structure called the quasiperiodic lattice structure [35]. Due to its special structure, I-phase could have many interesting properties such as high hardness, good thermal stability, high corrosion resistance, low coefficient of friction, and low interfacial energy [36,37]. By periodically introducing steps and ledges along the interfaces, the strong atomic bonding at the I-phase/matrix interfaces was achieved [38,39]. Previous work demonstrated that the formation of I-phase was beneficial to improving the mechanical strength both at room and elevated temperatures [18,21,40,41], thermal stability [10,20], and corrosion resistance of Mg-Li alloys [42,43]. Therefore, the strength improvement of Mg-14Li-6Zn-1Y alloy should be closely related to the existed I-phase particles.

For the strong textured metallic materials, the orientation factor (Ω) for the dislocation movement on a slip plane is inversely proportional to their yield strength and can be expressed as:

Ω = cos α·cos β (2)

where α and β are the angles between tensile direction and optimal slip direction and slip plane, respectively. Supposing that the angles of α and β are in the same plane, they should be complementary. For Mg-14Li alloy with BCC structure, the preferential slip plane is {110} planes and slip direction is <111> direction [44,45]. Based on the pole figures of Mg-14Li alloy (Fig. 4(a)), the angles between {110} planes and tensile direction of “ED” samples are 45°, whereas the corresponding angles of “TD” samples mainly concentrate at 90° ± 10°. Thus, for the Mg-14Li alloy, the Ω values of {110} planes in “ED” samples are 0.5, whereas the corresponding Ω values in “TD” samples are less than 0.17. Consequently, the “ED” samples have the better plasticity than that of the “TD” samples. Meanwhile, due to the soft orientation (i.e. easier activation of slips on {110} planes) of most grains in “ED” samples, their YS and UTS values are much lower than those of “TD” samples.

As for the Mg-14Li-6Zn-1Y alloy, its texture intensity is about 4.4 time lower than that of Mg-14Li alloy (Fig. 4). Following this, it can be predicted that the mechanical anisotropy of Mg-14Li-6Zn-1Y alloy should be remarkably reduced. Meanwhile, the existing texture causes the “ED” and “TD” samples respectively having soft and hard orientations, resulting in the higher plasticity of “ED” samples. However, it is interesting to be found that “ED” samples can also have higher tensile strength (Fig. 5). Thus, it can be deduced that besides the crystallographic texture, the distribution of I-phase particles should also be considered as a key factor for influencing the mechanical anisotropy of Mg-14Li-6Zn-1Y alloy. After extrusion, the broken I-phase particles were zonally distributed along “ED” (Fig. 2(a)). Previous work demonstrated that when tensile testing was performed perpendicular to the zonally distributed phase particles, the plasticity and tensile strength will be simultaneously degraded [46]. Although I-phase particles have the strengthening effect, their zonal distribution can easily induce the stress concentration [17]. To relieve the incompatible deformation occurring in I-phase and β-Li matrix, cracks will preferentially nucleate at the I-phase/β-Li interfaces when the tensile testing was performed along the “TD” direction (Fig. 7(h)). Due to the brittle nature of I-phase [49], micro cracks can also be occasionally observed in the interior of I-phase particles. With increasing the tensile strain, more and more cracks can be formed and coalesce with each other to induce the formation of big cracks. Thus, the zonal distribution of I-phase particles in Mg-14Li-6Zn-1Y alloy can accelerate the fracture process of “TD” samples, resulting in their simultaneous decrease in both tensile strength and elongation. For the HCP-structured Mg-4Li-6Zn-1.2Y alloy, the distribution of I-phase particles can also induce the nucleation of numerous micro cracks [21]. However, for the duplex structured Mg-6Li-6Zn-1.2Y alloy [22], the incompatible plastic deformation can easily occur at the α-Mg/β-Li interfaces because the β-Li phases have more independent slip systems thanα-Mg phases, which resulted in the preferential nucleation of micro cracks at α-Mg/β-Li interfaces. Based on the discussion mentioned above, it can be concluded that although the I-phase/β-Li interfaces is much stronger than α-Mg/β-Li interfaces, the high stress concentration at big-sized I-phase particles can remarkably reduce the resistance to the nucleation of micro cracks. Therefore, to effectively make use of the potential strengthening effect of I-phase particles in differently structured Mg-Li alloys, it is essential to control their sizes and distribution. In the future work, the target is to explore suitable processing methods for the refinement of I-phase particles and investigate their effect on the mechanical behavior.

4. Conclusion

Through investigating the effect of icosahedral phase on the crystallographic texture and mechanical anisotropy of BCC-structured Mg-14Li based alloys, the conclusions can be summarized as follows. The formed I-phase particles in Mg-14Li-6Zn-1Y alloy can be beneficial to the refinement of grain size and weakening of the texture intensity. Meanwhile, the I-phase formation can result in the improvement of mechanical strength. For the Mg-14Li alloy, “ED” samples have the higher tensile strength and lower elongation when compared with the “TD” samples, which is ascribed to the formed strong cube texture. Although the texture in Mg-14Li-6Zn-1Y alloy is weaker, the zonal distribution of I-phase particles causes the simultaneous degradation in tensile strength and elongation of “TD” samples.

Acknowledgements

This work was supported financially by the Strategic New Industry Development Special Foundation of Shenzhen (No. JCYJ20170306141749970), the National Natural Science Foundation of China (Nos. 51871211 and 51701129), the Natural Science Foundation of Guangdong Province (No. 2018A030313950), the funds of International Joint Laboratory for Light Alloys, the National Key Research and Development Program of China (Nos. 2017YFB0702001 and 2016YFB0301105), Liaoning BaiQianWan Talents Program, the Innovation Fund of Institute of Metal Research (IMR),Chinese Academy of Sciences (CAS).


/