Development of a High Strength Mg-9Li Alloy via Multi-Pass ECAP and Post-Rolling

Abstract

In this study, a high-strength Mg-9Li alloy was developed via multi-pass equal-channel-angular-pressing (ECAP) and post rolling, of which the yield tensile stress (YTS) and ultimate tensile stress (UTS) were 166 MPa and 174 MPa representing about 219% and 70% increase in YTS and UTS respectively, compared to the cast alloy. The cast alloy was ECAP processed at 200 °C for 4, 8, and 16 passes, followed by room-temperature rolling to a total thickness reduction of 50%. The 8-passes ECAPed (E8) alloy presented the best strength of all the ECAPed alloys, and the post rolling endowed the alloy (E8R) further strengthening and the best strength of all the alloys. Grain-boundary strengthening and dislocation strengthening were the two major factors for the high strength of the processed alloys. The α-Mg phase grains were greatly refined to about 2 μm after 8-passes ECAP, and was further refined to about 800 nm ~1.5 μm after rolling. Significant grain refinement endowed the alloy with sufficient grain-boundary strengthening. Profuse intragranular dislocation accumulated in the deformed matrix, leading to the significant dislocation hardening of the alloy. Rolling-induced strong basal texture of the α-Mg phase also enhanced the further strengthening of the E8R alloy.

1. Introduction

Magnesium (Mg) alloys have high specific strength and ductility resulting in their use in various applications from automobile, aerospace, and orthopedic applications. However, its low cold workability near room temperature due to its hexagonal packed crystal structure limits its use [1,2,3]. Magnesium alloys have been classified by many researchers as a super lightweight structural metal, attracting a lot of attention in the automobile and aerospace industries where production of light, super-fast machines requiring less fuel consumption through weight reduction is of high priority [4,5]. Magnesium–Lithium (Mg–Li) alloys are gaining more and more interest in scientific research studies as well as in industrial applications owing to their super lightweight, high specific strength and good formability making it one of the most ideal structural materials for 3C intelligent electronics, medical devices etc. [6,7,8].

Many researchers have employed the use of alloying and various mechanical procedures to enhance the mechanical property of Mg-Li base alloys [9,10]. Notable amongst them is the ECAP (equal channel angular pressing) process for achieving ultrafine grains (UFG) by severe plastic deformation (SPD) [11,12]. ECAP has advantages of producing large bulk materials for various industrial applications, with improvement in strength and ductility properties [13,14,15]. Conventional ECAP process however has typical drawbacks of reinserting billets into the die for every pass, resulting in inconsistencies in temperature applied for each pass [16,17,18]. Some researchers have therefore resorted to the development of novel processes with higher production efficiency, such as repetitive upsetting (RU) [19,20,21] and rotary die equal channel angular pressing (RD-ECAP) [22,23,24]. Also, another effective plastic deformation in use is the rolling technique. Magnesium alloys have a dense hexagonal structure with few slip systems. This makes it easy to induce stresses when rolling at large strain rates which results in very low yield of magnesium alloy sheets [25]. However, the addition of lithium to the magnesium alloy forming Mg–Li alloy results in the formation of a β-rich Li phase which greatly increases the dislocation slip system and improves the plastic deformation ability of the Mg-Li alloys [26]. The results obtained from the reported research [27,28] show that, the use of rolling technique leads to an improvement in the strength properties of the Mg-Li alloy, activate more slip systems, enhance the ability of intergranular co-opening and improve the ability of plastic deformation at room temperature (RT). In the rolling process, strong basal texture will be formed [29]. Research done by the reference [28] employed unidirectional (transverse and longitudinal) and cross (combination of transverse and longitudinal) rolling routes to improve the mechanical properties of Mg-9Li-1Al alloy. The unidirectional rolling routes reached 170 MPa tensile strength whereas the cross rolling route reached 243 MPa respectively. During the rolling process, compression twins may be formed. Therefore, in the rolling process, the original structure can be refined by pretreatment and the morphology and texture of the material can be controlled during the rolling process. Different researchers [30,31] have employed the use of ECAP with further rolling, applying it in a wide variety of alloys and have achieved sufficiently good tensile strength and ductility.

This research therefore seeks to investigate the effect of employing a combination of RD-ECAP with further room-temperature rolling techniques on the microstructural evolution and mechanical property changes of Mg-9Li duplex alloy.

2. Experimental

2.1. Materials and Processing

The raw material Mg-9Li alloy ingots used in the experiment was purchased from Jiangsu Li Mg Aero Material Co., Ltd. The composition of the alloy was analyzed by GNR S3 spark direct reading spectrometer. The results are shown in

Table 1. Analysis of Mg-9Li alloy composition (wt. %).

Mg	Li	Fe	Mn	Zn	Cd
91.052	8.809	0.010	0.024	0.014	0.030

. The raw material can be identified as Mg-9Li alloy according to the mass ratio of Mg to Li element. According to the binary phase diagram of Mg–Li alloy, the lithium content of the alloy in this paper falls in the range (between 5.7–10.3 wt. %) of the (α + β) duplex phase structure.

ECAP was conducted at 200 °C with the samples extruded for 4, 8, and 16 passes. The rotary Die equal channel angular pressing (RD-ECAP) set-up as diagrammatically illustrated in Figure 1a with detailed information obtained from our previous investigation [22,27] consists of a plunger for forcing the sample through the die orifice with all four sides of the die sealed with punches with the upper punches protruding out of the die to be pressed by the plunger. This is used for continuously processing the samples. This method is in tandem with the principles of the conventional ECAP with the extruded inner angle being 90°. The advantage over the conventional ECAP set-up is that unlike in the former, each pass needs to be rotated and then refilled with the mold and rotated clockwise at the end of each pass as shown in Figure 1a. The rotary die can save a lot of experiment time, shorten the holding time of each pass and effectively reduce the dynamic recovery behavior after long time heating [27]. The inner parts of the die is cleaned with graphite emulsion to reduce friction during the extrusion process. After each pass, the sample squeezes from the upper channel into the left channel. The mold is then rotated clockwise with the sample reverting to its initial upper channel position, which is ready for the next pass. The as-cast sample is heated to 200 °C for half an hour in an oven and then processed for 4, 8, and 16 passes respectively. The preparation of the 16-pass ECAP sample requires another half an hour reinsertion of the die into the furnace after 8 passes is completed before performing the further 8 passes. This is to ensure that the sample does not crack during the extrusion process.

The ECAPed specimens are further rolled at room temperature (RT) to obtain high strength and toughness. All the ECAP-processed specimens were further rolled along a 20 × 40 × 2.5 mm³ thin section perpendicular to the extruded surface using a rolling machine as diagrammatically illustrated in Figure 1b with detailed information obtained from references [27,31]. Multi-pass rolling at room temperature was used via the rolling speed of 0.1 mm∙s⁻¹ in this process. Each pass reduction was 10% until the total reduction reached 50%. This method can effectively prevent the sample from cracking during the rolling process. The height between the two rollers were adjusted with the speed controlled. After each pass, the sample is reversed for the next roll to ensure the uniformity of the rolling. Each sample is designated with a specific name. The as-cast alloy, ECAPed alloys with 4, 8, 16 passes, cast-rolled alloy, 4-passes ECAP plus rolling alloy, 8-passes ECAP plus rolling alloy and 16-passes ECAP plus rolling alloy are given designated names of C, E4, E8, E16, CR, E4R, E8R, and E16R respectively.

2.2. Microstructure Characterization

The microstructure of Mg-9Li dual phase alloy after ECAP and rolling was analyzed by Optical Microscopy (OM, Olympus BX51M, Tokyo, Japan). The samples were cut into 10 × 10 mm small squares, then grounded using sandpaper increasing the grit size after each grinding cycle. The samples were then polished using 0.5 µm Al₂O₃ suspension solution until a mirror-like specimen surface was observed. The polished samples were etched with a solution mixture of 4.3 mL picric acid, 95 mL ethanol, and 0.7 mL phosphoric acid.

Transmission electron microscopy (TEM, FEI Tecnai G2, Hillsboro, OR, USA) was applied to observe the microstructure and grain size of the Mg-9Li alloy after ECAP and post-rolling. The samples were cut into a 10 × 10 × 0.6 mm³ pieces. The cut piece was grounded using a 1000 and 2000 grit sandpaper to about 100 µm. The sample was then punched into circular disks of 3 mm diameter and a special auxiliary equipment was used to polish the disc to 20–30 µm on the 1500 grit sandpaper. The TEM foil was finally perforated by ion milling (Gatan 695C, Pleasanton, CA, USA).

The phase composition of Mg-9Li dual phase alloy in different states was qualitatively characterized by X-ray diffractometer with a Bruker D8 (XRD, Karlsruhe, Germany) with test conditions: Cu target Ka-ray, tube current 40 mA, scanning range 10–90°, scanning speed 5°/min. The texture of the samples in different states were observed by XRD. The samples were polished using a 1000 grit sandpaper to ensure a smooth surface and etched with 3% nitric acid to remove its surface stress with the direction of extrusion and rolling indicated on the sample surface.

2.3. Tensile Test

Tensile test of the Mg-9Li alloys were conducted using the uniaxial tensile test at room temperature via a UTM4204X electronic tensile machine (Suns, Shenzhen, China) with a strain rate of 1 × 10⁻³ s⁻¹. A dog bone shape of dimensions 6 × 2 × 2 mm³ was used with at least five samples tested for each process sample and the average value obtained. The mechanical properties were measured and the fracture morphologies of the different processed alloys were observed using a scanning electron microscope (FEI Quanta 3D FEG, Hillsboro, OR, USA).

3. Results

3.1. Microstructure of ECAPed and ECAP-Rolled Alloys

The optical microstructure of the cast alloy and ECAPed alloys with different processing passes are shown in Figure 2. As seen in Figure 2a, the grey phase of the cast alloy is indicative of the α-rich Mg phase and the darker phase is β-rich Li phase. Clearly, the α-Mg phase is large and irregularly distributed in the β-Li matrix. After ECAP processing, the microstructure of α-Mg phase is continuously deformed with increase in number of passes and the reduction in grain size accordingly. As shown in Figure 2b, after four passes, the α-Mg phase shows elongated phases distributed in a particular angle showing a distinct plastic flow pattern. In addition, compared with the as-cast microstructure, the β-Li phase also decreased after 4 passes ECAP. After ECAP for 8 passes, the α-Mg phase grains are further lengthened with both phases oriented in the direction of the extrusion flow lines, as shown in Figure 2c. After 16-passes ECAP, as shown in Figure 2d, the α-Mg phase is further refined, presenting the typical plastic-deformation flows.

Figure 3a shows the optical microstructure of cast-rolled Mg-9Li alloy in the rolling direction. Clearly, the α-Mg phase is elongated and arranged in order along the rolling direction. Figure 3b,c is the optical microstructure of the ECAP-rolled alloy observed from the rolling direction (RD) and normal to the rolling direction (ND) respectively. Note that, this alloy was firstly ECAP processed for 8 passes, and then subjected to rolling until a 50% reduction in thickness. Thus, the samples of Figure 3b,c are named as E8R-RD and E8R-ND, respectively. Compared to the ECAPed alloys, the duplex phases of the alloy was further elongated and arranged in order along the rolling direction, presenting the typical plastic deformation flow. Meanwhile, the alloy has a more severe plastic-deformation flow in the direction normal to the rolling direction, presenting the typical fibrous microstructure.

Figure 4 presents the XRD spectra of the Mg-9Li alloys after ECAP and post rolling. As shown in Figure 4a with the cast and ECAPed alloys with different passes, it is clear that the peaks, as well as their intensities, of the two-phase structure changes with the number of ECAP passes in the alloy. Compared with the as-cast alloy, the peaks of the α-Mg and β-Li phases have many different crystal faces. For example, (200), (220), (310) peaks are formed in the Li-rich phase whiles ($10\overline{1}2$), ($11\overline{2}0$) peaks are formed in the α-Mg phase. Comparatively, after 8 passes, the α-Mg shows peaks of $(10\overline{1}0),(10\overline{1}1),(11\overline{2}0)$ whereas the peak increased obviously and the diffraction peaks of (200) and (211) crystal planes are enhanced in the β-Li phase. The XRD spectra of the ECAP-rolled alloys are shown in Figure 4b. The diffraction peaks of the β-Li phase (110) crystal plane are weakened after rolling whereas the diffraction peaks are stronger on the basal (200) and cone (310) planes after rolling. Among all the diffraction peaks of the α-Mg phase, the diffraction peak of the basal plane (0002) is the strongest, which indicates the existence of strong basal texture in the ECAP-rolled alloys [32,33,34]. Meanwhile, the peaks of (200) crystal planes of the ECAPed alloys were further strengthened after rolling, and this phenomenon was more obvious in the alloys with larger ECAP passes.

3.2. Tensile Mechanical Properties of the ECAPed and ECAP-Rolled Alloy

The tensile engineering strain-stress curves of the alloy after different-passes ECAP process and post rolling were obtained and presented in Figure 5, and the related mechanical parameters summarized in

Table 2. Tensile test properties of as-cast, ECAP and ECAP-rolled Mg-9Li alloy.

Mechanical Properties	C	E4P	E8P	E16P	CR	E4R	E8R	E16R
UTS (MPa)	102	106	133	116	158	133	174	149
YTS (MPa)	52	88	110	100	152	126	166	120
Eu (%)	15	5	7	5	3	3	2	7
Ef (%)	33	25	24	31	16	21	22	26

. Due to the amount of soft β-Li phase of the duplex structure, the cast alloy has limited strength and excellent ductility, presenting the extremely low yield strength (YTS, about 52 MPa) and high elongation to failure (E_f, about 33%). One characteristic noteworthy is that, the cast alloy presents sufficient tension work-hardening ability, presenting the continuously strengthened tensile bearing capacity. Benefited from this advantage, the ultimate strength (UTS, about 102 MPa) of the cast alloy is nearly double of the YTS and also presents the excellent performance in the uniform elongation (E_u, about 15%).

ECAP process improved the strength but decreased the ductility of the Mg-9Li alloy, and the mechanical performance changed with different ECAP passes. After ECAP for 4 passes, the YTS and UTS of the E4 alloy were improved to 88 MPa and 106 MPa, and the E_u and E_f decreased to 5% and 25% respectively. With increased ECAP passes, the 8-passes ECAPed alloy (E8) was further greatly strengthened with better ductility compared to the E4 alloy, of which the YTS and UTS are about 110 MPa and 133 MPa, and the E_u and E_f are about 7% and 24%. However, remarkable softening of the alloy occurred when the ECAP process was further increased to 16 passes. Compared to the E8 alloy, the E16 alloy presented lower YTS (about 100 MPa) and UTS (about 116 MPa). However, the ductility of E16 alloy was improved, of which the E_f was about 31%, nearly the same as that of the cast alloy. However, it should be emphasized that the strength of the E16 alloy is still significantly higher than that of cast alloy, especially the YTS. Different to the sufficient work-hardening ability of the cast alloy during the tensile process, the ECAPed alloys presented a reduced difference between values of YTS and UTS, indicating their less tension work-hardening ability.

Figure 5b presents the engineering strain-stress curves of the rolled alloys. Different to the tensile curves of the cast and ECAPed alloys, the tensile curves of the rolled sample showed the obvious dense serrated fluctuations after yielding which can be adjudged to be the C-type Portevin-Le Chatelier (PLC) effect [9]. The appearance of this phenomenon may be caused by the instability of the Mg atoms in the β-Li matrix after rolling. When plastic deformation occurs under the action of tensile stress, dislocation passes through these unstable solid-solution Mg atoms, which promotes the dissolution of these Mg atoms from the β-Li matrix. At the same time, the dislocation will be pinned at the grain boundary or the dislocation accumulation during the movement, so the interaction between the movable dislocation and the unstable dissolved Mg atom will cause the fluctuation of the tensile curve [9,35]. Significant strengthening and dramatic reduction in ductility occurred to the cast alloy after rolling process, of which the YTS and UTS increased to about 152 and 158 MPa, and the E_u and E_f decreased to about 3% and 16% respectively. Judging from the typical change in strength and ductility of the cast-rolled alloy, one can point out that strain-induced dislocation hardening is the major factor dominating the above phenomenon. During the rolling process, a large number of dislocations are induced within the grains. With further deformation, the edge dislocations and screw dislocations move along (0002) planes. Until the grain boundary is reached, a large number of dislocations in the grain boundary tangles further from the dislocation cell walls, further impeding the slip of dislocations.

Post rolling also improved the strength and decreased the ductility of the ECAPed alloys. However, all the ECAP-rolled alloys had better ductility compared to the cast-rolled alloy. Among them, the E8R alloy had the best strength with satisfactory ductility, of which the YTS, UTS, and E_f values reached 166 MPa, 174 MPa and 22% respectively. Compared to the cast alloy, the E8R alloy dramatically enhanced the YTS and UTS, which increased by 219% and 70% respectively. Compared to the cast-rolled alloy, the YTS and UTS increased by 10% and 9%. The most important thing is that, the E8R still kept satisfactory ductility after rolling while the cast-rolled alloy suffered dramatic reduction in ductility. Different to the E8 alloy, the rolling process endowed a quite limited strengthening to the E4 and E16 alloys, of which the YTS and UTS were all less than that of the CR alloy. This phenomenon may have a close relationship to the incompletely refined microstructure of the E4 alloy and the softening process of the E16 alloy.

Figure 6 presents the SEM fracture morphology of Mg-9Li alloys. Generally, as seen in Figure 6a, the cast alloy has a typical tearing fracture, indicating the excellent ductility. The fracture of the as-cast alloy shows major deep dimples in the β-Li phase induced by further growth of the voids due to the accumulation of dislocations and the few quasi-cleavage fracture along the α-Mg phase. In the E8 alloy (Figure 6b), the fracture surface is relatively flat with fewer marks of severe tearing compared to the cast alloy, indicating less plasticity. In addition, the fracture is majorly occupied by a mass of small shallow dimples while a few cleavage steps can be also observed. The great decrease in dimple size should be closely related to the grain refinement. Post rolling created more obvious flat fracture to both the cast and ECAPed alloy, indicating remarkable reduced ductility. As seen in Figure 6c, the CR alloy has an equal amount of dimples, mixed with large cleavage steps. Meanwhile, its dimples are much shallower than that of the cast alloy. As seen in Figure 6d, the E8R alloy also presents increased cleavages compared to the E8 alloy. However, the cleavage of this alloy was obviously less than that of the CR alloy, indicating the satisfactory and better ductility. From the above analysis, it can be concluded that the SEM fracture morphologies properly reflect the alternating ductility of the alloys after ECAP and post rolling.

4. Discussion

4.1. Grain-Boundary Strengthening and Dislocation Strengthening

Generally, thermoplastic processing greatly influences the mechanical properties of a metal from three aspects [36,37,38]. Firstly, the process eliminates the casting defects of the raw materials. Secondly, the coarse grains will be greatly refined due to the strain-induced dislocation evolution, resulting in the significant grain refinement. Thirdly, the coupling effect of heat and strain-induced dislocation entanglement will stimulate the dynamic recrystallization of the deformed matrix. For the Mg-9Li alloy, due to the bcc structure and its softer essence of β-Li phase, it will be firstly deformed with lots of dislocations accumulated in the α/β phase interface, which inhibits the recovery of the β-Li phase but stores up enormous distortion energy [39]. With increase in ECAP passes and induced strain, a large number of dislocations will concentrate inside the deformed β-Li phase grains [26,40]. The increase in dislocation density leads to the sub-grain boundary becoming the nucleation point of recrystallization, eventually growing up to from new grains [18,41]. Therefore, dynamic recrystallization takes place preferentially in β-Li phase, which may lead to a limited strengthening and the potential softening of the alloy [42].

As revealed in the tensile curves, the E4 alloy has an improved YTS compared to the cast alloy with limited UTS and decreased ductility. The dynamic-recrystallization of β-Li phase and the limited grain refinement of the α-Mg phase is the major contributing factor to this phenomenon as well as the extremely limited further strengthening to the E4 alloy after further rolling. As revealed in the tensile curves, both E8 and E8R alloys presented the best strength, due to the combination of grain-boundary strengthening and dislocation strengthening from both α-Mg and β-Li phases. Generally believed, the finer the grain size during thermal plastic deformation, the more easily dynamic recrystallization occurs [40,43]. Due to the re-heating of the E16 alloy after 8-passes followed by an additional 8 passes, a high possibility of intensive dynamic recrystallization occurring to both α-Mg phase and β-Li phase due to the lattice distortion energy obtained from the further ECAP-induced plastic deformation results in the typical softening phenomenon of the E16 alloy.

TEM images of the E8 and E8R alloys are presented in Figure 7. Figure 7a presents the typical grain morphologies of deformed α-Mg phase. The selected electron diffraction pattern (SAED) in the bottom right corner of Figure 7a was observed in the ([$41\overline{5}3$]) zone axis. As marked, the crystal planes of $(1\overline{1}10),(1102),\mathrm{and}(\overline{1}013)$ clearly shows the typical lattice feature of the Mg phase. After continuous ECAP process for 8 passes, obvious grain refinement of approximately 2 µm has been achieved in the deformed α-Mg phase. As shown in Figure 7b, the deformed Mg grains has been further refined to about 800 nm to 1.5 μm with improved grain boundary strengthening, satisfying the Hall-petch equation, stated as σ_y = σ₀ + kd^−1/2 [44]. Figure 7c presents the typical intragranular dislocations of the E8R alloy. Based on the TEM microstructure analysis, it can be deduced that the strength enhancement of E8 and E8R alloys was dominated by the combined effect of sufficient grain-boundary strengthening and dislocation strengthening with additional induced straining at room temperature and the obtained finer grains. A number of researches have shown that the intensity and plasticity of the α-Mg phase in the hcp structure can be improved synchronously after severe grain refinement via severe plastic deformation (SPD) process [45]. Also, according to the reported results of the SPDed UFG Mg alloys [46,47], the great refinement of the Mg grains will benefit both high strength and high ductility of the α-Mg phase of the alloy. However, the large plastic deformation also leads to the grain refinement, as well as the strain-induced work hardening of the β-Li phase, leading to the observed decrease in ductility of the E8 alloy.

4.2. Texture Strengthening

As reported by many researches [26,48], texture also greatly influences the strength and ductility of the Mg alloys. As revealed in the XRD patterns of the alloys, the $\left(10\overline{1}0\right),\left(10\overline{1}1\right),\mathrm{and}\left(11\overline{2}0\right)$ peaks of the ECAPed alloys were strengthened. Meanwhile, the strongest diffraction peak of the basal plane (0002) was detected. Thus, one can infer that the texture may play an important role in the mechanical property of the processed Mg-9Li alloy. The XRD texture of the E8 and E8R alloys are presented in Figure 8. Due to the cumulative effect of shear stress during the ECAP deformation, the basal texture of the α-rich Mg phase is 40–60° with the ECAP-extrusion direction, and the maximum texture density is up to 6.1. Prismatic texture $\left(10\overline{1}0\right)$ and conical texture $\left(10\overline{1}1\right)\mathrm{and}\left(11\overline{2}0\right)$ are found with weak texture density. Generally, during plastic deformation of magnesium alloys, the slip in the direction of the basal plane is the first thing to occur [16]. Due to the addition of Li, the ratio of c/a crystal axis is decreased, which is favorable to the dislocation slip in the cylindrical and conical directions. In conjunction with the orientation distribution function (ODF) diagram in Figure 8b, it is found that it is not only the basal plane, prismatic, and cone textures that appear in the α-Mg phase but also there exists ($11\overline{2}0$) <$1\overline{1}00$> textures. The texture shows that the local grains were ordered after ECAP processing since the grains in different directions are arranged differently by shear force after multi-pass ECAP processing. These multi-directionally ordered textures block the slip of the grains, leading to an increase in the strength. Figure 8c depicts the pole figures for the {0002} plane of the α-Mg phase, a strong basal texture along the rolling direction can be seen, and it is the appearance of this texture that leads to the significant increase in strength of the E8R alloy. The distribution of the $\left\{10\overline{1}0\right\}$ prismatic weak structure in the rolling direction is 15°. The $\left\{10\overline{1}1\right\},\left\{11\overline{2}0\right\}$ conical planes have no particularly apparent texture.

5. Conclusions

In summary, a combined process of multi-pass ECAP and post rolling technologies were employed in the development of a high-strength Mg-9Li duplex alloy. The microstructure and their influence on the strength and ductility of the processed Mg-9Li alloys were systematically investigated. The main conclusions are:

(1)

Cast Mg-9Li alloy was firstly processed via multi-pass ECAP at 200 °C for 4, 8, and 16 passes to achieve grain refinement in both α-Mg and β-Li phases of the ECAPed alloys. Post rolling was conducted at room temperature to obtain further strengthening of the alloys. All the alloys after the combined process presented enhanced strength and decreased ductility compared to the cast alloy.

(2)

Among all the ECAPed alloys, the E8 alloys presented the best strength, of which the YTS and UTS are 110 MPa and 133 MPa, respectively. Post rolling of the E8 alloy further strengthened the alloy and endowed it with the best strength of all the alloys in this research. The YTS and UTS of the E8R alloys reached 166 MPa and 174 MPa. Approximately 219% and 70% increase in YTS and UTS was achieved compared to the cast alloy, respectively.

(3)

Grain-boundary strengthening and dislocation strengthening are the key factors to the greatly improved strength of the Mg-9Li alloys after the combined processing. Significant grain refinement of the α-Mg phase was achieved in the E8 alloy, of which the grain size was about 2 μm. Post rolling further reduced the grain size to between 800 nm and 1.5 μm. With the greatly refined grains, the grain-boundary strengthening of the E8 and E8R alloys was obtained. Profuse intragranular dislocation was accumulated in the deformed matrix of the E8 and E8R alloys, leading to the significant dislocation hardening of the alloy.

(4)

Prismatic texture $\left(10\overline{1}0\right)$ and conical texture $\left(10\overline{1}1\right)\mathrm{and}\left(11\overline{2}0\right)$ were detected with weak texture density in the E8 alloy. The strong basal texture along {0002} plane of the α-Mg phase was formed in the rolled Mg-9Li alloys along the rolling direction, which also contributed to the most improved strength of the E8R alloy.