Influence of rolling directions on microstructure, mechanical properties and anisotropy of Mg-5Li-1Al-0.5Y alloy

(1)

Full Length Article

Influence of rolling directions on microstructure, mechanical properties and

anisotropy of Mg-5Li-1Al-0.5Y alloy

Tianzi Wang

a

, Tianlong Zhu

b

, Jianfeng Sun

c

, Ruizhi Wu

a,

*

, Milin Zhang

a

a_{Key Laboratory of Superlight Materials & Surface Technology, Ministry of Education, Harbin Engineering University, Harbin 150001, PR China} b_{Shenyang Jinbo Gas Compressor Manufacturing Co. Ltd., Shenyang 110027, PR China}

c_{School of Materials Science and Engineering, Heilongjiang University of Science and Technology, Harbin 150022, PR China}

Received 10 October 2015; accepted 9 November 2015 Available online 2 December 2015

Abstract

Mg-5Li-1Al-0.5Y alloy was rolled with different directions. The microstructure, mechanical properties and texture of the specimens were investigated with optical microscope, tensile tester and X-ray diffraction. The results show that changing rolling directions can refine the grain size of as-rolled alloys. Meanwhile, rolling directions have an obvious influence on the mechanical properties and texture of Mg-5Li-1Al-0.5Y alloy, thus affecting the anisotropy of the alloy. The sheet, of which the RD (rolling direction) and ND (normal direction) are both changed between two passes, possesses the smallest anisotropy. From the texture results, changing rolling directions reduces the maximum pole density, making the highest point distribution region excursion and the highest point distributes more scatteredly.

Keywords: Mg-Li alloy; Rolling directions; Microstructure; Mechanical property; Anisotropy

1. Introduction

As the lightest metallic engineering material with low density, high specific stiffness and strength, good magnetic shielding and shock resistance ability, Mg-Li alloy is a prom-ising material in the fields of aerospace, military, automotive and electronic industry[1,2]. However, because of the relatively coarse grains and severe inhomogeneity microstructure, the as-cast Mg-Li alloy always possesses a poor strength, which limits its application.

The work performed by Herbstein and Averbach indicated that the addition of lithium decreases the crystal lattice axial ratio (c/a) of magnesium[3]. Aluminum is the commonly used alloying element that can improve the mechanical properties of Mg-Li alloy[4]. Trace rare-earth (RE) element can significantly weaken the basal texture of the Mg-Li alloy and increase its ductility[5,6]. Yttrium is one of the typical RE element used in the Mg-Li alloy.

To improve the strength and ductility of the alloy, alloying and plastic processing are two effective ways. In previous work, the influence of RE on microstructure and mechanical proper-ties of Mg-Li based alloy has been studied in detail[7,8]. The studies about plastic processing of Mg-Li alloy mainly concen-trate on the influence of single deformation or rolling deforma-tion with single direcdeforma-tion [9–11]. However, there is rare literature about the influence of rolling directions on micro-structure, texture and mechanical properties of Mg-Li alloys.

In this paper, the microstructure, mechanical properties and anisotropy of the Mg-5Li-1Al-0.5Y alloy rolled with three dif-ferent rolling directions were comparatively investigated. The texture evolution and plastic deformation mechanism of the alloy were also discussed.

2. Materials and methods

The Mg-5Li-1Al-0.5Y alloy used in the experiment was prepared from CP (commercial pure) magnesium (99.9%), CP lithium (99.9%), CP aluminum (99.9%) and the master alloy of Mg-17wt.%Y. The materials were melted in a medium fre-quency induction furnace under the protection of argon gas. The melt was cast into a book-shape steel mold, then the ingot was cut into specimens with dimensions of 120 mm× 100 mm× 5 mm. Before rolling, the specimens were annealed

* Corresponding author. College of Materials Science & Chemical Engineering, Harbin Engineering University, 145 Nantong Street, Harbin, 150001, PR China. Tel.:+86 451 82569890; fax: +86 451 82569890.

E-mail addresses: [email protected], [email protected] (R. Wu).

http://dx.doi.org/10.1016/j.jma.2015.11.001

Available online atwww.sciencedirect.com

Journal of Magnesium and Alloys 3 (2015) 345–351 www.elsevier.com/journals/journal-of-magnesium-and-alloys/2213-9567

(2)

at 573K for 12h for homogenization and stress relieving. The rolling process was carried out at a temperature of 373K through two-pass rolling schedule with a total cumulative reduction of 34%. The rolling schedule is listed inTable 1. The

rolling directions between rolling passes are shown inFig. 1. Deformation 1 means that RD (rolling direction), TD (trans-verse direction) and ND (normal direction) are all invariable. In this process, the sheet is named as sheet 1. As for deformation 2, RD deflects 180 °C, while TD and ND are invariable. The corresponding sheet is named as sheet 2. As for deformation 3, TD is invariable, while RD and ND deflect 180 °C. The corre-sponding sheet is named as sheet 3. At the interval between passes, the specimens were annealed at 373K for 0.5h. After rolling, the specimens were annealed at 523K for 0.5h.

Specimens for microstructure observation were polished and etched with a solution of 6 vol.% nital mixed with a proper

Table 1

Rolling schedule of the alloy. Passes Entrance thickness (mm) Export thickness (mm) Rolling reduction (mm) Reduction rate (%) 1 5.00 4.00 1.00 20.00 2 4.00 3.30 0.70 17.50

Fig. 1. Diagrammatic drawing of different rolling processes: (a) deformation 1, (b) deformation 2 and (c) deformation 3.

(3)

amount of picric acid. Then the microstructure was analyzed with LEICA DM IRM optical microscope. The strength and elongation were tested with MTS 50KN tensile tester at room temperature under an initial rate of 1× 10−3_s−1_{. The gauge}

dimensions of tensile specimen were 3 mm× 3 mm × 20 mm. The texture of the rolled specimens was measured using X‘ pert PRO X-ray diffractometer, analyzing incomplete pole figures between 5 °C and 70 °C with back reflection mode using Cu-Kα radiation at 40kv and 40 mA. The pole figures were recorded using the rotation of {0002},

{

10 01

}

,

{

10 11

}

and

10 21

{

}

reflection, then the inverse pole figures were calculated through the four reflections.

3. Results

3.1. Microstructure

Fig. 2 shows the microstructure of as-cast and as-rolled Mg-5Li-1Al-0.5Y alloys. The grain size of the as-cast alloy is the largest among the four specimens, and some black second-ary phases exist in the alloy. Zhu et al.[12] reported that the black secondary phase is Al2Y. Compared with Fig. 2a, the

grain size of the as-rolled specimens is much smaller (shown in

Fig. 2b, c and d); sheet 3 possesses the smallest and compara-tively more uniform grain size. The Al2Y compounds still exist

in the specimens, but most of them are crushed into fine par-ticles. Few bulk Al2Y phase could also be found in sheet 1 and

sheet 2. While in sheet 3, only some tiny particulate phase distributes in the grains.Fig. 3is the microstructure of sheet 1 before annealing. Many twins apparently distribute in grains. Compared with Fig. 2, after annealing treatment, there exists almost no twins in the grains.

3.2. Mechanical properties

Table 2lists the strength and elongation of the specimens. The data show that, after rolling, the tensile strength increases obviously, while the elongation slightly decreases.Fig. 4shows the tensile strength and yield strength variations of as-rolled Mg-5Li-1Al-0.5Y with three different rolling directions. Along RD, sheet 1 and sheet 2 possess the similar values ofσbandσ0.2,

somewhat higher than that of sheet 3. Along TD, the three sheets possess the similarσ0.2values, while sheet 3 possesses a

higherσbthan the others.Fig. 5shows the elongation variation

of the as-rolled alloys along RD and TD, respectively. With the changing of rolling directions, the elongation of the sheet slightly decreases. The three sheets all show an inconspicuous plastic anisotropy along TD and RD.

Combined withTable 2, Fig. 4and Fig. 5, it can be con-cluded that the strength along RD is higher than that along TD. The tensile strength along RD of the three as-rolled sheets shows little difference. Along TD, the tensile strength value of the three specimens are lower than that along RD. The tensile

Fig. 3. Optical micrograph of sheet 1 (before annealing).

Table 2

The strength and elongation of Mg-5Li-1Al-0.5Y alloy.

Strength (MPa) Elongation (%)

RD TD RD TD σ0.2 σb σ0.2 σb As-cast alloy 52 125 52 125 24.6 24.6 Sheet 1 75 154 68 139 23.1 23.3 Sheet 2 77 156 65 138 22.2 22.0 Sheet 3 68 151 64 150 21.8 21.9 0 20 40 60 80 100 120 140 160 Stress/MPa Specimens σ0.2 σb

Sheet 1 Sheet 2 Sheet3

a

0 20 40 60 80 100 120 140 160 St re e s /MPa Specimens σ0.2 σb

Sheet 1 Sheet 2 Sheet 3

b

(4)

strength of sheet 3 is higher than that of sheet 1 and sheet 2, almost the same as that along RD. The elongation value of the three sheets all keeps a high level, and changing rolling direc-tions shows no significant impact on it. Sheet 3 possesses almost no anisotropy.

3.3. Texture

Figs. 6–9illustrate the texture of the three as-rolled speci-mens.Fig. 6is the pole figures of sheet 1. In the pole figure of

(0002), the grain orientation is located approximately 35 °C from the c-axis towards TD. There also exists highest intensity points distributing about 55 °C from the c-axis toward RD in the

(

10 11

)

pole figure, and approximately 45 °C from the c-axis toward downside of RD in the

(

10 21

)

figure. The pole density distribution is scattered in the pole figure of

(

10 01

)

. Compared with the pole figures of other magnesium alloys with basal texture, it can be inferred that a non-basal texture exists in the alloy during rolling.

0 5 10 15 20 25 30 E longati on/% Specimens

a

0 5 10 15 20 25 30 Elon ga tion /% Specimens

b

Fig. 5. Elongation of as-rolled Mg-5Li-1Al-0.5Y alloys: (a) RD and (b) TD.

(5)

As for sheet 2, the maximum texture intensity distributes about 45 °C from the c-axis toward RD in the (0002) pole figure; the pole figures of

(

10 01

)

,

(

10 11

)

and

(

10 21

)

do not exist in a large pole density region (as shown inFig. 7).Fig. 8

is the pole figures of sheet 3. The basal peak distributes about 55 °C from the c-axis toward the intermediate between RD and TD. As the same with sheet 2, there exists no large pole density region in the pole figures of

(

10 01

)

,

(

10 11

)

and

(

10 21

)

. However, as for sheet 2 and sheet 3, the texture intensity varies in a narrow range, which means weak grain-oriented.

Fig. 9 reveals the inverse pole figures of the as-rolled Mg-5Li-1Al-0.5Y alloys. In these images, RD is the macro-scopic coordinate direction. InFig. 9a, the crystal orientation concentrates on the 10 01 direction, namely 10 01 // RD. In

Fig. 9b, the crystal orientation gradually excursion to the

11 02 direction, and it distributes more scatteredly than sheet 1. InFig. 9c, the maximum intensity of the texture distributes more scatteredly between 10 01 and 11 02 . It indicates that changing rolling direction weakens the texture of as-rolled alloy, increasing the random of the texture. Meanwhile, with the changing of rolling direction, the maximum of pole density decreases from 1.545 to 1.303, thus keeping the elongation of the alloy remains at a high level.

From the results mentioned above, it can be concluded that the as-rolled specimens do not possess a typical basal texture, showing that the non-basal slip is activated during rolling at 373K. ComparingFigs. 6–8, with the changing of rolling direc-tions, the highest intensity region distribution gradually moves outwards from the c-axis and the scope of the texture intensity becomes narrow. Fig. 9 illustrates that, after changing the rolling directions, the pole density decreases. Meanwhile, the texture region gradually deflects from 10 01 to 11 02 , until it evenly distributes. The changing of rolling directions weakens the anisotropy of the alloy, and sheet 3 possesses outstanding comprehensive mechanical properties.

4. Discussion

Due to the low stacking fault energy of Mg-Li alloy, rolling deformation of the alloy is given priority to DRX (dynamic recrystallization). Ma et al.[13]shows that new grains gener-ated by DRX with a low dislocation density also participate in the texture evolution as the strain increases.

In this study, the rolling temperature is 373K, which is lower than the recrystallization temperature. However, alloys contain-ing rare earth elements and/or yttrium (RE/Y) can develop more random textures [14]. The addition of Y causes the

(6)

formation of Al2Y phase. The particles of Al2Y compound in

magnesium alloys can promote DRX during the process of rolling and provide more approaches for nucleation of DRX texture, weakening the basal texture. PSN (particle stimulated nucleation) is the reason of weakening the basal texture. In

addition, the rare earth elements atoms that solute in magne-sium alloy will not only change the bond energy of RE-Mg and Mg-Mg around rare earth atoms, but also change the stacking fault energy of basal and non-basal planes, thus changing CRSS (critical shear stress) of the basal and non-basal slip systems,

Fig. 8. Pole figures of sheet 3.

(7)

narrowing extended dislocation and causing it being easily trapped into perfect dislocation[15]. Therefore, the non-basal slip system can be activated easily.

After two-pass rolling, some twins are found in the grains. The stress relief annealing makes many of them disappear, instead of isometric crystal grains, causing the grain refine-ment. Three types of rolling deformation processes make dif-ferent influences on the alloy. As for the sheet 1, the upside surface suffers a relatively larger deformation than the down-side surface; meanwhile, the stress inhomogeneously distrib-utes on the same surface, and DRX easily occurs in the coarse grains. After changing the rolling directions, such as sheet 2, the same surface possesses similar deflection, increasing the uniformity of grain size. Sheet 3 not only possesses similar deflection in the same surface, but also makes the upside and downside surfaces be deformed uniformly. The grains are refined and become more uniform with the changing of rolling direction, which weakens the anisotropy of the alloy. Sheet 3 possesses the smallest grain size and the most uniform grains, thus possessing the outstanding comprehensive mechanical performance.

As a supplement coordinate deformation mechanism, the twining during rolling process can not only provide some inde-pendent slip systems, but also cut the coarse grains and cause grain refinement [13]. Changing rolling directions can affect the distribution of accumulated deformation twinning, espe-cially the compression twinning and double twinning. During the low temperature rolling and heat treatment, the recrystalli-zation occurs. With the procedure of transformation from twin boundaries into recrystallization grain boundaries, the texture evolves with a small deflection, thereby affecting the strength and elongation of the alloy. The anisotropy of the alloy also changes.

Kaiser et al.[16]proposed that the distribution of the (0002) basal plane has a significant influence on the planar anisotropy of the sheet. As mentioned by Chen et al. [17], the texture exhibits greater spreading of the basal poles toward the RD than the TD, whereby activation of the basal slip in the sheet rolling direction is preferred to the TD. This would result in lower stresses necessary for the 90 °C orientation, hence decreasing the yield strength. Lee et al. [18] calculated the Lankford parameter of the rolled RD, TD and ND directions. The results revealed that the r-values of the RD plane are much greater than those of the TD and ND planes. Therefore, the strength in RD is larger than that in TD. With the changing of rolling directions, the basal pole in (0002) basal plane starts deflecting, no longer toward RD. As for sheet 3, the basal peak distributes from the c-axis toward the intermediate between RD and TD. Mean-while, the maximum of the pole density decreases in the inverse pole figures, thus weakening the anisotropy of the alloy.

5. Conclusions

The rolling directions affect the microstructure and texture of Mg-5Li-1Al-0.5Y alloy, thus making influences on the mechanical properties of the alloy.

1 The changing of rolling directions can refine the grains. 2 The changing of rolling directions keeps the strength and

elongation remaining at a high level both along RD and TD. Rolling directions affect the anisotropy of the alloy. The sheet, of which the RD and ND are both changed between two passes, has the smallest anisotropy.

3 The changing of rolling directions affects the texture of Mg-5Li-1Al-0.5Y alloy. It reduces the maximum pole density, making the highest point distribution region excur-sion and the highest point distributes more scatteredly. 4 The addition of yttrium promotes DRX through PSN

mecha-nism, weakening the basal texture and reducing the anisot-ropy of the alloy.

Acknowledgements

This work was supported by the Heilongjiang Province Youth Skeleton Program (1252G018), Research Fund for the Doctoral Program of Higher Education (20132304110006), Project of Science and Technology of Heilongjiang Province Education Department (12511068), Fundamental Research Funds for the Central Universities (HEUCF20151006), Heilongjiang Province Natural Science Foundation (E201420) and Harbin City Application Technology Research and Devolopment Project (2015AE005).

References

[1] D.K. Xu, E.H. Han, Scr. Mater. 71 (2014) 21–24.

[2] R.Z. Wu, Y.D. Yan, G.X. Wang, L.E. Murr, W. Han, Z.W. Zhang, et al., Inter. Mater. Rev. 60 (2015) 65–100.

[3] F.H. Herbstein, B.L. Averbach, Acta Metall. 4 (1956) 407–413.

[4] R.Z. Wu, Z.K. Qu, M.L. Zhang, Rev. Adv. Mater. Sci. 24 (2010) 35–43.

[5] M. Li, H. Hao, A.M. Zhang, Y.D. Song, X.G. Zhang, J. Rare Earths 30 (2012) 492–496.

[6] N. Stanford, D. Atwell, A. Beer, C. Davies, M.R. Barnett, Scr. Mater. 59 (2008) 772–775.

[7] X.Y. Guo, R.Z. Wu, J.H. Zhang, B. Liu, M.L. Zhang, Mater. Des. 53 (2014) 528–533.

[8] L. Bao, Q.C. Le, Z.Q. Zhang, J.Z. Cui, Q.X. Li, J. Magnes. Alloys 1 (2013) 139–144.

[9] H.W. Dong, F.S. Pan, B. Jiang, Y. Zeng, Mater. Des. 57 (2014) 121–127.

[10] D. Zdeneˇk, T. Zuzanka, K. Stanislav, J. Alloys Compd. 378 (2004) 192–195.

[11] M. Karami, R. Mahmudi, J. Mater. Sci. 44 (2013) 3934–3946.

[12] T.L. Zhu, J.F. Sun, C.L. Cui, R.Z. Wu, B. Sergey, Z. Leng, et al., Mater. Sci. Eng. A 600 (2014) 1–7.

[13] Q. Ma, B. Li, W.R. Whittington, A.L. Oppedal, P.T. Wang, M.F. Horstemeyer, Acta Mater. 67 (2014) 102–115.

[14] J. Bohlen, M.R. Nürnberg, J.W. Senn, D. Letzig, S.R. Agnew, Acta Metall. 55 (2007) 2101–2112.

[15] C.L. Cui, L.B. Wu, R.Z. Wu, J.H. Zhang, M.L. Zhang, J. Alloys Compd. 509 (2011) 9045–9049.

[16] F. Kaiser, D. Letzig, J. Bohlen, A. Styczynski, C.H. Hartig, K.U. Kainer, Mater. Sci. Forum 419-422 (2003) 315–320.

[17] X.P. Chen, D. Shang, R. Xiao, G.J. Huang, Q. Liu, Trans. Nonferrous Met. Soc. China 20 (2010) s589–s593.

[18] B.H. Lee, S.H. Park, S.G. Hong, K.T. Park, C.S. Lee, Mater. Sci. Eng. A 528 (2011) 1162–1172.