Investigation on the Microstructure and Mechanical Properties

of Mg-Al-Yb Alloys

Su Mi Jo1, Kyung Chul Park1, Byeong Ho Kim1, Hisamichi Kimura2,Sung Kyun Park3;* and Yong Ho Park1;*

1Department of Material Science and Engineering, Pusan National University, Busan 609-735, Korea2Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan3Department of Physics, Pusan National University, Busan 609-735, Korea

The effects of Yb addition on the microstructure and mechanical properties of Mg-5Al alloy are investigated. The results indicate that theaddition of Yb to the Mg-5Al alloy facilitates the formation of a thermally stable Al2Yb phase, the refinement of the microstructure and thesuppression of the volume fraction of Mg17Al12 phase in Mg-5Al alloy. Yb addition has little effect on the mechanical properties of theexperimental alloys tested at room temperature. At elevated temperatures, however, the ultimate tensile strength (UTS) is significantly increasedby Yb addition and Mg-5Al-1Yb has the highest UTS value than other experimental alloys. On the other hand, the yield strength (YS) increasesat all tested temperatures due to the grain refinement and dispersion strengthening of the secondary phase. Meanwhile, the elongation (") of theexperimental alloys decreases at all tested temperatures. Tensile fractographic analysis indicates that cleavage fracture is the dominantmechanism of the Mg-5Al and Mg-5Al-xYb alloys at room temperature. At elevated temperatures, however, the fracture mechanism ofexperimental alloys mainly changes from cleavage to quasi-cleavage fracture. [doi:10.2320/matertrans.MC201011]

(Received December 1, 2010; Accepted April 7, 2011; Published May 25, 2011)

Keywords: magnesium alloys, magnesium-alminum-rare earth, ytterbium addition, mechanical property

1. Introduction

As one of the lightest structural materials, magnesium andits alloys have great potential for applications in automotive,aerospace and other industries due to their low density, highspecific strength and good damping capacity.1–3) Mg-Alseries alloys, such as AZ91 and AM60, are widely used dueto their high castability and wide range of room temperaturemechanical properties.4) However, the applications of thesealloys are limited at elevated temperatures due to their poorcreep resistance, which is typically improved by formingthermally stable precipitates to prevent grain boundarysliding during creep deformation. The most effective alloyingelements for such purposes are rare earth metals (RE) thatsignificantly improve the creep-resistance.5)

Although some creep-resistant Mg-Al alloys with REhave already been developed, not many investigations havebeen reported in the literature concerning the effect ofRE addition on the mechanical properties of Mg alloys.RE are normally added to Mg alloys such as misch metal(MM) because MM has a similar behavior with that of RE,but the cost is lower than RE. In spite of its lower price,some studies reported that the mechanical properties ofsingle RE-added Mg alloys are higher than those of thealloys with added MM. Earlier studies investigated somesingle RE such as La, Ce, Nd and Sm.6–10) However, noresearch about the effects of ytterbium (Yb) addition on themicrostructure and mechanical properties of Mg alloys hasbeen reported.

In this investigation, the effects of Yb addition on themicrostructure and mechanical properties of Mg-5Al alloyare studied in order to provide a reference for the develop-ment of new RE-containing Mg alloys.

2. Experimental Procedure

The Mg-5Al-xYb (x ¼ 0, 0.5, 1 and 2 mass%) alloys withnominal compositions listed in Table 1 were prepared usinghigh purity magnesium (99.9%), aluminum (99.99%) andMg-40 mass%Yb master alloy. The alloys were melted at750�C in a magnesia crucible under a CO2 + SF6 gasatmosphere and poured into a preheated permanent mould at200�C. Microstructural analysis was carried out using anoptical microscope (OM) and a scanning electron microscope(SEM) equipped with an energy dispersive X-ray spectrom-eter (EDS). A solution of an acetic picral (10 mL aceticacid + 4.2 g picric acid + 10 mL distilled water + 70 mLethanol (95%)) was used to etch the samples. The samplesselected at the corresponding position were homogenized at420�C for 4 h in order to delineate the grain. The phases inthe as-cast alloys were analyzed by X-ray diffraction (XRD)using monochromatic CuK� radiation. The volume fractionsof the precipitates were measured using an image analyzer.The dimensions of the tensile test specimens were2 mm� 2 mm� 13:2 mm. The tensile tests were carriedout at an initial engineering strain rate of 3:33� 10�2 s�1 atroom and elevated temperatures (150�C, 200�C). An offsetmethod is used to determine the yield strength of the material

Table 1 Nominal compositions of as-cast Mg-5Al alloys with and without

Yb addition (mass%).

Alloy Al Yb Mg

Mg-5Al — Bal.

Mg-5Al-0.5Yb 5 0.5 Bal.

Mg-5Al-1Yb 5 1 Bal.

Mg-5Al-2Yb 5 2 Bal.

*Corresponding author, E-mail: psk@pusan.ac.kr, yhpark@pusan.ac.kr

Materials Transactions, Vol. 52, No. 6 (2011) pp. 1088 to 1095Special Issue on Platform Science and Technology for Advanced Magnesium Alloys, V#2011 The Japan Institute of Metals

tested. The values of �0:2 achieved by the strain gage. SEMwas used for the fractographic observations to clarify thefracture process.

3. Results and Discussion

3.1 MicrostructureFigure 1 shows the XRD results of the as-cast experimen-

tal alloys. The XRD patterns revealed the main phases of theMg-5Al alloy to be �-Mg and Mg17Al12. With increasing Ybcontent, the peak of the Al2Yb phase was newly appeared.

The optical microstructures of the four studied alloysare shown in Fig. 2. Some phases were precipitated in the

Mg-5Al alloy (Fig. 2(a)), and were revealed by the XRDresults to be Mg17Al12 phase. With Yb addition, otherphases appeared at the grain boundaries in the Mg-5Al-xYb alloys. (Fig. 2(b)–(d)) Mapping analysis was performedto determine the elemental distribution of Mg, Al and Ybin the microstructure shown in Fig. 3. Al and Yb elementsappeared around the grain boundaries, which indicated thatMg17Al12 and Al2Yb phases had precipitated at the grainboundaries.

Figure 4 shows the SEM images and EDS-spectra of theexperimental alloys. According to the XRD and EDS results,the only secondary phase precipitated at the grain boundarieswas Mg17Al12 in the Mg-5Al alloy (Fig. 4(a)) and Al2Ybphase was newly formed at the grain boundaries in the Mg-5Al-xYb alloys (Fig. 4(b)–(d)). The morphology of Al2Ybphase was changed to a more clearly lamellar structure andbecame finely dispersive at the grain boundary with increas-ing Yb content. However, at a Yb content of 2 mass%, Al2Ybphase was coarsened (Fig. 4(d)). The volume fractions of theMg17Al12 and Al2Yb phases were decreased and increased,respectively, with increasing Yb content. According to L.Ke-jie et al.,9) the electronegativity difference between twoelements and the solidification kinetics of a metallic melt canbe used to predict the possibility of intermetallic formation.The larger the electronegativity differences between the twoelements, the stronger the bond and the higher the possibilityof intermetallic formation. As the electronegativity differ-ence between Al and Yb is higher than that between Al andMg, the formation of Al-Yb intermetallics is favored overthat of Mg-Al intermetallics in the Mg-Al-Yb system, andMg17Al12 phases are suppressed by Yb addition.10,11)

Fig. 1 X-ray diffraction patterns of AZ51 alloys with and without Yb

addition.

Fig. 2 Optical micrographs of the experimental alloys: (a) Mg-5Al, (b) Mg-5Al-0.5Yb, (c) Mg-5Al-1Yb and (d) Mg-5Al-2Yb alloy.

Investigation on the Microstructure and Mechanical Properties of Mg-Al-Yb Alloys 1089

Figure 5 shows the microstructures of the experimentalalloys after homogenized treatment. Homogenized treat-ment at 420�C for 4 h ensured the complete dissolution ofthe Mg17Al12 phase, which is evident from the micro-structure of the homogenized Mg-5Al alloy shown inFig. 5(a). Figure 5(b)–(d) shows the microstructures ofhomogenized Mg-5Al-0.5, 1 and 2Yb alloys, respectively.As already shown in Fig. 5(a), there was no Mg17Al12

phase, and only Al2Yb phase still appeared at the grainboundaries even after 4 h of homogenized treatment, whichindicated that the Al2Yb phase was thermally more stablethan the Mg17Al12 phase.12) In addition, the grain size wasgradually decreased with increasing Yb content in the Mg-5Al. J. Zhang et al.13) have reported that based on theprinciples of solidification and the binary phase diagram ofthe Mg-RE system, the distribution coefficient of soluteRE is less than 1, so that during the solidification process,solute atoms of RE, as well as Al, are enriched in theliquid ahead of the solid–liquid interface. This may lead toconstitutional undercooling and a reduced diffusion rateof atoms, thereby increasing the number of nuclei andrestricting the grain growth. On the other hand, theenrichment of solute atoms leads to the formation of Al-RE phases, which are mainly distributed in the grainboundary, which further inhibits the grain growth. As anRE, therefore, Yb addition seems to induce a certain effecton grain refinement.

3.2 Mechanical properties3.2.1 Tensile behavior

The tensile properties of ultimate tensile strength (UTS),yield strength (YS) and elongation (") were tested from roomtemperature to 200�C and the results are shown in Fig. 6.Figure 6(a) shows the tensile properties of the alloys testedat room temperature. Compared to Mg-5Al alloy, althoughthe difference was very slight, Mg-5Al-1Yb had the highestUTS. Meanwhile, the YS of the alloys gradually increasedand " slightly decreased with increasing Yb content. Theelevated temperature tensile properties are shown in Fig. 6(b)and (c). Yb addition improved UTS and YS, but decreased" at elevated temperatures in Fig. 6(b) and (c). Mg-5Al-1Yballoy had the highest UTS than other experimental alloys atboth elevated temperatures and room temperature. The Mg-5Al-1Yb alloy was the optimized composition among theexperimental alloys at room and elevated temperatures.

At room temperature, the slight UTS variation among thefour alloys was attributed to the similar room temperatureproperty of the Mg17Al12 and Al2Yb phases. At elevatedtemperatures, however, the increased tensile strength with Ybaddition was attributed to the thermal stability of the Al2Ybphase. The main strengthening phase of the Mg-5Al alloy isMg17Al12, which has a low melting point (approximately437�C) and poor thermal stability, and hence can readily besoftened at temperatures exceeding 120–130�C. However,upon Yb addition to the Mg-5Al alloy, Al2Yb phase formed

Fig. 3 Mapping analysis of alloying elements on the surface of Mg-5Al-0.5Yb alloy.

1090 S. M. Jo et al.

at the grain boundary. Al2Yb, which has a high melting pointof approximately 980�C, has great thermal stability, andtherefore acted to increase the UTS in the 1 mass% Yb addedalloy. However, when Yb addition exceeded 2 mass%, theAl2Yb phase was coarsened. The presence of massivesecondary phase causes stress concentration, and intergranu-lar cracks are readily formed when the alloy is loaded.14)

Therefore, the UTS of the Mg-5Al-2Yb alloy was slightlydecreased at elevated temperatures. Meanwhile, YS increas-ed in all tensile tests at different temperatures. In general, YSis related to grain size. According to the well-known Hall–Petch relationship, grain refinement can obviously increasethe YS.10) As shown in Fig. 5, grain size was efficientlyrefined with increasing Yb content. The " was decreased withincreasing Yb content at room and elevated temperatures.This was attributed to the brittleness of Al2Yb, which has ahigher thermal stability but is rather brittle.3.2.2 Fracture behavior

The SEM images of the tensile fracture surfaces of theexperimental alloys at room temperature and 200�C areshown in Figs. 7 and 8. In general, cleavage fracture, quasi-cleavage fracture and intergranular fracture are known to be

the main fracture modes of magnesium alloys. Figure 7(a)shows fracture surface of the Mg-5Al alloys tested at roomtemperature. The fracture surface reveals a cleavage planewith some secondary cracks. In the Mg-5Al alloy, Mg17Al12

phase was the only secondary phase and the Mg17Al12 phaseswere discontinuously precipitated and very susceptible tofracture because of their brittleness.15) The presence of manycleavage planes with cleavage steps of different sizes andriver patterns indicated the action of brittle failure. With Ybaddition, more and larger cleavage surfaces were observedin Fig. 7(b)–(d), which was attributed to the decrease inductility with increasing Yb content. At elevated temper-atures, the fracture mode was changed from cleavage toquasi-cleavage, as evidence in the fracture surface by thepresence of numerous plastic zones spread over the entirefracture surface, as seen in Fig. 8.16) However, some cleavageplanes with secondary cracks remained along with thedeformation zone, which indicated a brittle fracture thatdiffered from cleavage. This is a rather complicated fracturepattern according to Y. Lu et al.17) The planes on the fracturesurface were incoherent with the Mg matrix but were formedthrough a combination of locally formed microcracks and

Fig. 4 SEM images and EDS spectra of Mg-5Al alloys: (a) Mg-5Al, (b) Mg-5Al-0.5Yb, (c) Mg-5Al-1Yb and (d) Mg-5Al-2Yb alloy.

Investigation on the Microstructure and Mechanical Properties of Mg-Al-Yb Alloys 1091

Fig. 5 Microstructures of the alloys homogenized treatment at 420�C for 4 h: (a) Mg-5Al, (b) Mg-5Al-0.5Yb, (c) Mg-5Al-1Yb and

(d) Mg-5Al-2Yb alloy.

Fig. 6 Tensile properties of the experimental alloys at (a) 25�C, (b) 150�C and (c) 200�C.

1092 S. M. Jo et al.

Fig. 7 Tensile fractograph of the experimental alloys at room temperature: (a) Mg-5Al, (b) Mg-5Al-0.5Yb, (c) Mg-5Al-1Yb and (d) Mg-

5Al-2Yb alloy.

Fig. 8 Tensile fractograph of the experimental alloys at 200�C: (a) Mg-5Al, (b) Mg-5Al-0.5Yb, (c) Mg-5Al-1Yb and (d) Mg-5Al-2Yb

alloy.

Investigation on the Microstructure and Mechanical Properties of Mg-Al-Yb Alloys 1093

simultaneously formed tearing ridges. The bottoms of the pitswere not strict cleavage planes but rather consisted of severalsomewhat sunken planes with secondary cracks. Figure 9shows a locally high magnification image of Fig. 8(c)that presents the tearing ridges and secondary crack. Thisindicated that Yb addition led to quasi-cleavage-type failureof the alloys at elevated temperatures.

In summary, the improved mechanical properties of Mg-5Al-xYb alloys at elevated temperatures were based on thefollowing two main strengthening effects.

First is the grain refinement effect. As shown in Fig. 5,with increasing Yb content, the grain size of the Mg-5Alalloys was decreased. The larger surface volume of the grainboundary effectively acted as a dislocation barrier, whichincreased the deformation resistance.8) According to theHall–Petch relationship grain refinement can obviouslyincrease the YS of Mg alloys due to the increased value ofthe coefficient k value.

�0:2 ¼ �0 þ k=ffiffiffi

dp

ð1Þ

where �0:2 is the YS, �0 a constant, d the average graindiameter, and for Mg k ¼ 280{320 MPa

ffiffiffiffiffiffiffi

mmp

,18) the valueof k was chosen as 300 MPa

ffiffiffiffiffiffiffi

mmp

. And the values for �0

(35 MPa) and d (590 mm) of pure Mg by permanent moldcasting were used in this calculation.19) The quantitative grainsize of the experimental alloys needed for calculating YS(�0:2) shows in Table 2. The YS (�0:2) could be estimatedusing the Hall–Petch relationship, as seen in Table 3. Asshown in Table 3, the measured �0:2 was higher than thecalculated �0:2. This indicated the action of another effectin improving the YS of the alloys, in addition to the grainrefinement effect.

This second effect is the dispersion strengthening. The heatresistance of the Mg alloys was determined by the softeningresistance of the �-Mg, the morphology, and the size anddistribution of the second phases in the alloy. With increasingYb content, Mg17Al12 phase and thermally stable Al2Ybphase were decreased and increased, respectively (Table 4),

and Al2Yb phase was finely dispersive at the grain boundary(Fig. 4). Therefore, the Mg-5Al-1Yb alloys, which hada finer grain size and more uniform distribution of grainboundary phases, had the highest mechanical properties thanthe other alloys. For the Mg-5Al-2Yb alloy, the decrease ofthe gap between the measured YS and the YS calculated bythe Hall–Petch relationship can be understand in the samecontext (Table 3).

In conclusion, Yb addition can further improve the tensileproperties of Mg-Al-RE alloys for the two aforementionedreasons.

4. Conclusions

The microstructure and mechanical properties of Mg-5Al-xYb (x ¼ 0, 0.5, 1 and 2 mass%) alloys were investigated.The main results were as follows:(1) The Mg-5Al alloy consisted of �-Mg and Mg17Al12

phases. Upon Yb addition, Al2Yb phase was observedwith a gradually reduced Mg17Al12 phase. The grainsize was also refined with increasing Yb content in theMg-5Al alloy.

(2) Among the experimental alloys, the UTS values of theMg-5Al-1Yb alloy were the greatest than other alloys atboth room and elevated temperatures due to the grainrefinement and dispersion strengthening.

(3) At Yb addition less than 2 mass%, both grain refinementand dispersion strengthening affected the UTS results,which improved with increasing Yb content. at a Ybcontent of 2 mass%, although the grain was morerefined, secondary phases were coarsened at the grainboundary and the UTS of Mg-5Al-2Yb was decreasedat room and elevated temperatures. Meanwhile, theelongation (") of the experimental alloys decreased atall tested temperatures.

(4) Cleavage fracture was dominant mechanism of Mg-5Aland Mg-5Al-xYb alloys tested at room temperature.Elevated temperatures and Yb addition both induced achange in the failure mechanism of the alloys to quasi-cleavage fracture.

Fig. 9 High-magnification of Mg-5Al-1Yb at 200�C.

Table 2 Quantitative grain size of the experimental alloys.

Alloy Mg-5Al Mg-5Al-0.5Yb Mg-5Al-1Yb Mg-5Al-2Yb

Grain size (mm) 98.5 63.8 47.9 24.3

Table 3 The comparison of yield strength (�0:2) between the calculated

values using Hall–Petch relationship and measured ones.

Alloy�0:2

Calculated Measured

Mg-5Al 52.9 83

Mg-5Al-0.5Yb 60.2 88

Mg-5Al-1Yb 66 89

Mg-5Al-2Yb 83.5 90

Table 4 Quantitative volume fractions of secondary phases of the

experimental alloys.

Alloy Mg-5Al Mg-5Al-0.5Yb Mg-5Al-1Yb Mg-5Al-2Yb

Volume fraction of

Mg17Al12 (%)3.25 2.32 1.46 0.27

Volume fraction of

Al2Yb (%)— 1.08 2.12 4.43

1094 S. M. Jo et al.

Acknowledgements

This research was supported by a grant from theFundamental R&D Program for Core Technology of Materi-als, from the Ministry of Knowledge Economy, Republicof Korea and a grant-in-aid for support from the TohokuUniversity Global COE program.

REFERENCES

1) B. H. Kim, K. C. Park, Y. H. Park and I. M. Park: Mater. Lett. 65 (2011)

122–125.

2) M. Yang and F. Pan: Mater. Design 31 (2010) 68–75.

3) W. Qudong, C. Wenzhou, Z. Xiaoqin, L. Yizhen, D. Wenjiang, Z.

Yanging and X. Xiaoping: J. Mater. Sci. 36 (2001) 3035–3040.

4) A. A. Luo: Int. Mater. Rev. 49 (2004) 13–30.

5) K. C. Park, B. H. Kim, H. Kimura, Y. H. Park and I. M. Park: Mater.

Trans. 51 (2010) 472–476.

6) J. Zhang, M. Zhang, J. Meng, R. Wu and D. Tang: Mater. Sci. Eng. A

527 (2010) 2527–2537.

7) J. Zhang, P. Yu, K. Liu, D. Fang, D. Tang and J. Meng: Mater. Design

30 (2009) 2372–2378.

8) B. H. Kim, K. C. Park, Y. H. Park and I. M. Park: Mater. Sci. Eng. A

527 (2010) 6372–6377.

9) L. Ke-jie, L. Quan-an, J. Xiao-tian, C. Jun, Z. Xing-yuan and Z. Qing:

Scr. Mater. 60 (2009) 1101–1104.

10) J. Wang, J. Yang, Y. Wu, H. Zhang and L. Wang: Mater. Sci. Eng. A

472 (2008) 332–337.

11) T. Rzychon, A. Kielbus, J. Cwajna and J. Mizera: Mater. Charact. 60

(2009) 1107–1113.

12) A. Srinivasan, U. T. S. Pillai and B. C. Pai: Mater. Sci. Eng. A 527

(2010) 6543–6550.

13) J. Zhang, D. Zhang, Z. Tian, J. Wang, K. Liu, H. Lu, D. Tang and J.

Meng: Mater. Sci. Eng. A 489 (2008) 113–119.

14) P. Song, X. Guo and X. Jing: Spec. Cast. Nonfer. Alloy. 27 (2007)

88–91.

15) W. Xiao, S. Jia, J. Wang, Y. Wu and L. Wang: Mater. Sci. Eng. A 474

(2008) 317–322.

16) Y. Lu, Q. Wang, X. Zeng, W. Ding, C. Zhai and Y. Zhu: Mater. Sci.

Eng. A 278 (2000) 66–76.

17) A. Luo and M. O. Perquleryuz: J. Mater. Sci. 29 (1994) 5259–5271.

18) J. F. Nie: Scr. Mater. 48 (2003) 1009–1015.

19) A. A. Luo: Int. Mater. Rev. 49 (2004) 13–30.

Investigation on the Microstructure and Mechanical Properties of Mg-Al-Yb Alloys 1095