Creep and Corrosion Properties of the Extruded
Magnesium Alloy Containing Rare Earth
Chao-Chi Jain
*1and Chun-Hao Koo
*2Department of Materials Science and Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei, Taiwan 106, Republic of China
Effects of microstructures on the creep and corrosion properties were investigated in the Mg-8Al alloys with addition of the rare earth elements (La-rich Mischmetal, RE). The addition of RE to Mg-8Al alloy may form a stable intermetallic phase, Al11RE3 at elevated temperature, and suppress thephase (Mg17Al12) with poor stability at high temperature. The corrosion rate of the alloy slightly decreases with increasing the added RE contents. The constant-load creep behavior was examined at 423, 448 and 473 K under stresses between 40 and 100 MPa. The RE-containing alloy showed a prolonged period of stead-state creep compared to the Mg-8Al base alloy and slightly reduced minimum creep rates. A stress exponent of 2 estimated suggests that the creep behavior can be controlled by the grain boundary sliding. The apparent activation energy for creep of Mg-8Al and Mg-8Al-2RE alloys are 114 and 104 kJ mol1, respectively.
[doi:10.2320/matertrans.48.265]
(Received July 6, 2006; Accepted December 21, 2006; Published January 25, 2007)
Keywords: magnesium-aluminum alloy, microstructure, creep property, corrosion property
1. Introduction
The challenge for producing high-efficient vehicles is promoted by a growing demand for the automotive industry to reduce air pollution and energy consumption. Magnesium alloy has advantages of the lowest density, and the excellent specific strength and stiffness among the commercially available structural alloys. The use of magnesium would cause a weight saving of around 33% or 77% comparing with aluminum alloys or iron/steel, respectively. Therefore, a magnesium-base alloy is usually a structural candidate for mass-saving applications.1–3) Mg-Al-Zn series alloy,
espe-cially AZ91 alloy, is the most popular magnesium alloy, since the AZ91 exhibits excellent die castability and a good balance of strength and ductility.4–6)However, the
applica-tions of AZ91 alloy have limited due to the low strength, poor creep and corrosion resistance at temperatures in excess of 393 K.7) The deterioration of the properties makes it
unsuitable for many components in automobile engines.8,9) The previous investigation10)showed that the grain boundary sliding is significant to the deformation mechanism at elevated temperatures in Mg-Al based alloys. Therefore, to improve heat resistance in Mg-Al based alloy, it seems an effective way to develop an alloy containing stable phases in thermal condition, which suppress the grain boundary sliding. Although some magnesium alloys containing Ca element that exhibit favorable mechanical properties, even at moderate temperature, developed in the past few years11–15)
but the castability is inferior to Mg-Al-Zn alloy. The addition of rare earth (RE) elements effectively works to improve the mechanical properties of magnesium alloys at elevated temperatures as well as the corrosion resistance.16,17) The
rare earth, used as Mischmetal (Mm, a mixture of rare earth elements) in the magnesium alloy, is a better choice to protect from both creep and corrosion viewpoints at elevated
temperature, since commercially Mm is much cheaper than individual rare earth element metals. The Mg-8Al-xRE (x¼ 0, 1, 2 or 3 mass%) alloys were studied thoroughly, and show also good behavior on their tensile properties at either ambient or elevated temperatures; and even excellent super-plasticity through our previous investigations.18,19)Therefore, the present investigation aims at clarifying the effects of a small amount of RE on the microstructure and the creep resistance of the Mg-8Al alloy at the elevated temperatures up to 473 K; accordingly the corrosion properties of
Mg-8Al-xRE alloys were also studied.
2. Experiments
The alloys of Mg-8Al-xRE (x¼0, 1, 2 or 3 mass%) were prepared by melting and casting in the vacuum induction furnace under the protection of argon gas, and then extruded to a sheet 3 mm thick at 633 K. The extrusion ratio was90:1. The chemical compositions of the prepared alloys were analyzed by an inductivity coupled plasma atomic emission spectrometer (ICP-AES), and the result of the analysis is listed in Table 1. RE elements were added as Mischmetal with a composition (by mass) of 85.3% La, 11.6% Ce, 2% Pr, 0.3% Nd, 0.03% Fe and 0.02% Si. After the processing, the average grain size, d (d¼1:74L, L is the linear intercept size), was about 20mm. The alloys were examined by optical and scanning electron microscope (SEM) techniques to determine their microstructural features. X-ray diffraction (XRD) techniques, was applied to identify the phases from the crystallographic feature in this study. SEM was per-formed using a Philips XL-30 SEM, and XRD was conducted using Philips 1710 diffractometer with CuK emission. Solartron 1287 Electrochemical Impedance Spectroscopy were used for electrochemical corrosion tests. Before the corrosion test, the specimens were ground, polished, cleaned, and weighted. Immersion test of corrosion was conducted submerging in water solution with 3.5 mass% NaCl for 6 h. After the immersion, the corroded specimens were dried and
*1Graduate Student, National Taiwan University
*2Corresponding author, E-mail: chkoo@ccms.ntu.edu.tw
weighted; and then compared with the original specimens to calculate the weight loss precisely.
Creep tests were carried out under a constant-load using a creep machine equipped with a three zone furnace (manu-factured by Applied Test System Inc., Butler, PA). The extension of the specimen was measured by using a LVDT mounted on the sample at a constant temperature. The specimens for the creep test is flat, were cut from the extruded alloy sheets parallel to the extruded direction and had a cross section of 4mm3mm and 16 mm in gauge length. The creep tests were carried out at three temperatures of 423, 448 and 473 K. Temperature fluctuation was 2 K under inde-pendently applied constant-load between 40 and 100 MPa.
3. Results and Discussion
3.1 Microstructural analysis and phase identifications
The microstructure of the as-cast Mg-8Al alloy showed a typical dendritic structure with the primary-Mg phase, the eutectic phase(coarse Mg17Al12 precipitates) and the fine
lamellar (discontinuously precipitated Mg17Al12 ()) as
shown in Fig. 1(a). The microstructure was formed by a divorced eutectic reaction.20)The fine lamellae (discontinu-ously precipitated) phase was formed from the outer region of -Mg dendrites, and is supersaturated with Al below the eutectic temperature.21)When the RE was added in the
Mg-8Al alloy, the Al11RE3intermetallic phase was precipitated.
Figure 1(b) depicts a microstructure of the as-cast Mg-8Al-2RE alloy, which contains a needle-like intermetallic Al11RE3 phase and a nodule-like phase. With larger
amount of RE addition in the alloy both the amount and the coarseness of the Al11RE3 phase particles were more
increased, whereas the amount of phase became less and the phase turned into finer particles.18) Because the
intermetallic Al11RE3phase consumes most of the aluminum
atoms. The sizes of all the precipitates in the as-cast alloys are over 10mmand some exceed 20mm.
[image:2.595.45.549.84.149.2]The as-cast Mg-8Al-xRE alloy possibly has casting-defects, i.e., the as-cast alloy has pores/segregation, as seen in Fig. 1(b). The mechanical property of as-cast alloy is
Table 1 Chemical composition of the investigated alloys (mass%).
Nominal alloy Mg Al Zn Mn La Ce Pr Nd Total (RE)
Mg-8Al Remainder 7.78 0.62 0.11 — — — —
Mg-8Al-1RE Remainder 8.18 0.60 0.12 0.97 0.136 0.023 0.003 1.132
Mg-8Al-2RE Remainder 8.26 0.66 0.12 1.86 0.261 0.043 0.007 2.171
Mg-8Al-3RE Remainder 8.34 0.65 0.11 2.71 0.379 0.063 0.009 3.161
Contents Pr and Nd were calculated from the measured La content and the mischmetal composition.
(a)
Mg17Al12
α-Mg
Al11RE3
Mg17Al12
α-Mg
(c)
(d)
α-Mg
fine discontinuous containing alternate
lamellae of α-Mg and Mg17Al12phase
the coarse intergranular
Mg17Al12precipitates
(b)
Mg17Al12
α-Mg
Al11RE3
pore
[image:2.595.99.497.184.489.2]usually inferior to wrought alloy due to the defects. The ingots were hot extrusion to remove the cast-defects and enhance mechanical property. Consequently, the
Mg-8Al-xRE alloys used in present study were in as-extruded alloys. After hot extrusion, as shown in Figs. 1(c) and (d), all the precipitates were crushed to form homogeneously distribut-ed fine particles with size less than 10mm. All the as-extruded specimens exhibit a uniform equiaxed grain structure with the mean grain size of about 20mm. The morphology of the Mg-8Al alloy is an -Mg matrix with irregular -phase precipitates distributed inside grains and also along grain boundaries as shown in Fig. 1(c). Fig-ure 1(d) represents the Mg-8Al-2RE alloy, which containing the rod-like Al11RE3 intermetallic phase and
homogene-ously distributed. The sharp contrast of the Al11RE3
intermetallic phase in the BSE (backscattered electron) image apparently indicates the presence of the rare earth elements, due to the heavy element displays sharp contrast in BSE image (see Figs. 1(b) and (d)).
Figure 2 is the X-ray diffraction (XRD) patterns of the as-extruded Mg-8Al and Mg-8Al-2RE alloys. The peaks with high intensities are mainly corresponding to two phases, -Mg and(Mg17Al12), both of which in Mg-8Al and
Mg-8Al-2RE alloys. The structure of phase is a body-centered cubic.22) The Al
11RE3 precipitate has a body-centered
orthorhombic structure23,24) was present only in
Mg-8Al-2RE alloy. Figure 3(a) shows the backscattered electron image of the Mg-8Al-2RE alloy. The X-ray mappings of Mg, Al and La in Figs. 3(b)–(d) clearly present the concentrations of magnesium, aluminum and rare earth elements in both the
and Al11RE3 intermetallic compounds. No other
RE-containing intermetallics except Al11RE3 precipitates have
been observed in the present experiment. A series of research which was done by Wei et al.,21,24,25) established a stable Al11RE3; and even to make a complete Al11RE3 compound.
3.2 Corrosion tests
In corrosion tests, salt water was used as the corrosive
solution; so that potentiodynamic polarization and immersion tests were conducted in water containing 3.5 mass% NaCl. Figure 4 shows results of the potentiodynamic polarization of the alloy obtained in a typical corrosion test. The results based on the Tafel extrapolarization show that the corrosion current density (in A/cm2) of the Mg-8Al-xRE alloys decreases with the RE content, as shown in Table 2. Thus, the corrosion rate of Mg-8Al base alloy is higher than the alloy with RE element. The corrosion rates of the alloy slightly decrease with increasing the added RE contents.
Figure 5 shows the surfaces of the Mg-8Al-xRE alloys after immersion. During the period of immersion, no odorous gas was generated from the alloy surface. The corrosion of magnesium in salt water is based on corrosion reaction equation (1),26) according to which magnesium simply
displaces hydrogen out of the water, producing hydrogen gas. Mgþ2Hþ¼Mg2þþH
2 ð1Þ
Therefore, no odorous gas was hydrogen gas. Residues in gray color were observed on the surfaces of the
Mg-8Al-xRE alloys after corrosion tests. The relationship between rate of the weight loss and RE contents in the Mg-8Al-xRE alloys is shown in Fig. 6. The weight loss of the alloys was measured after the immersion in the 3.5 mass% NaCl solution for 6 h. The weight loss rate was reduced from 6.02 for Mg-8Al to 5.51 (in mg/cm2/day) for Mg-8Al-3RE
alloy. In the corrosion process, the trend of the rate determined from the weight-loss measurements was con-sistent with that determined by the electrochemical testing method (shown in Fig. 4 and Table 2). The experimental results in Fig. 6 show that Mischmetals containing such as La, Ce, Pr and Nd improve the corrosion resistance of the Mg-Al alloys in the chloride solutions. The solubility of rare earth in the Mg-Al alloys is limited by the presence of aluminum; but the intermetallic Al-RE phase formed during the immersion tests is electrochemical passive and does not affect the corrosion rate.27,28)The high corrosion resistance of the Mg-8Al alloy with RE is related to a certain positive
20 30 40 50 60 70 80 90
10
two theta, 2θ/θ/degree
Intensity
Mg17Al12 Mg
20 30 40 50 60 70 80 90
10
two theta, 2θ/θ/degree
Intensity
Mg Mg17Al12 Al11RE3
(a)
(b)
[image:3.595.103.501.71.294.2]synergism of aluminum atoms and RE elements in the Mg-8Al base alloy; consequently, impeding the propagation of the localized corrosion attack, as shown in Fig. 7. The
improvement of the Mg-8Al alloy by the added RE under the corrosion condition is possibly due to the enrichment of the trace elements of RE in the oxide film, and the benefits of Al11RE3 distributed uniformly in the matrix. The
(a)
(c)
(d)
(b)
Fig. 3 (a) backscattered electron image and X-ray maps of (b) Mg, (c) Al, (d) La of as-extruded Mg-8Al-2RE alloy obtained by energy dispersive X-ray spectroscopy.
I / A-cm-2
E / V
[image:4.595.99.497.72.367.2]Fig. 4 Potentiodynamic polarization curves with the 3.5 mass% NaCl solution for the Mg-8Al-xRE alloys.
Table 2 Potentiodynamic corrosion test data of the Mg-8Al-xRE alloys.
Alloy Corrosion potential E (Voltage)
Corrosion current density I (A/cm2)
Mg-8Al 1:2812 1:4638107
Mg-8Al-1RE 1:2536 1:3527107
Mg-8Al-2RE 1:2101 1:2139107
Mg-8Al-3RE 1:103 1:0871107
10 mm
[image:4.595.306.549.418.691.2] [image:4.595.49.289.421.586.2] [image:4.595.47.290.652.726.2]theoretical and detailed analysis is not clear yet, and the further investigation is under way.
3.3 Creep tests
3.3.1 Creep curves of the investigated alloys
Creep tests were conducted at three temperatures of 423, 448 and 473 K under applied constant-load between 40 MPa and 100 MPa. Table 3 lists the creep data for all the alloys tested under 40 MPa. At 473 K, the creep rupture life of the
Mg-8Al base alloy was only 14 h, while that the alloy with 3 mass% RE extended the life to 32 h. The creep rupture life of the RE-containing alloy may raises sharply with the RE addition i.e. even over 700 h at 423 K/40 MPa and 30 h at 473 K/40 MPa, for Mg-8Al-3RE. The creep rate of Mg-8Al is poor, which is as high as 2:51106s1 at 473 K/
40 MPa. However, the creep rate of Mg-8Al-3RE alloy is reduced to 6:77107s1 with 3 mass% of RE addition,
near one order of magnitude lower than that of Mg-8Al alloy. Figure 8 shows several creep strain vs. creep rupture life curves obtained typically from the constant-load and con-stant-temperature of Mg-8Al-xRE alloys under 40 MPa at (a) 473 K and (b) 423 K, respectively. These curves show a standard form of the constant-load creep test: where the primary, secondary (or steady-state), and tertiary stages were represented. There are three apparent regions can be distinguished at 423 K, especially for the Mg-8Al-2RE and Mg-8Al-3RE alloys (see Fig. 8(b)). However, at a higher temperature in our investigation, 473 K, the steady-state stage of the creep curve is unclear or even vanish, namely from primary stage immediate to get into tertiary stage until to fractured (see Fig. 8(a)). Adding 1 mass% RE slightly increased the creep rupture life of Mg-8Al alloy. However, the creep rupture life of Mg-8Al-2RE and Mg-8Al-3RE alloy significantly improved (see Fig. 8(b)). The results indicate that the creep resistance of the Mg-8Al-xRE alloys exceeds that of the Mg-8Al base alloy. The addition of the RE elements to the Mg-Al alloy could slightly reduce the steady-state creep rate and prolonged the creep rupture life. The creep rupture life at 423 K was improved twenty to thirty times more than that at 473 K, as shown in Table 3.
The main strengthening phase in the Mg-8Al alloy is the
phase, which has a low melting point of approximately
0 1 2 3
La-Rich Mischmetal Content, CLa/mass% 5.4
5.6 5.8 6 6.2
5.5 5.7 5.9 6.1
W
e
ight loss rate (mg/cm
2/day)
[image:5.595.55.284.72.296.2]Fig. 6 Average rates of weight loss of the Mg-8Al-xRE alloys change with the various RE contents after the immersion in 3.5 mass% NaCl solution for 6 h.
Table 3 Creep Properties of the Alloys Tested at 423, 448 and 473 K three temperatures, respectively under the applied stress of 40 MPa.
Alloy
423 K, 40 MPa 448 K, 40 MPa 473 K, 40 MPa
Creep life (h)
Steady-State creep rate (1/s)
Creep life (h)
Steady-State creep rate (1/s)
Creep life (h)
Steady-State creep rate (1/s)
Mg-8Al 354 7:36108 43 4:43107 14 2:51106
Mg-8Al-1RE 372 6:93108 112 2:06107 15 1:43106
Mg-8Al-2RE 564 3:51108 116 1:64107 19 8:67107
Mg-8Al-3RE 768 1:35108 167 1:29107 32 6:77107
(a)
(b)
[image:5.595.97.498.487.636.2] [image:5.595.47.548.695.784.2]735 K29)and poor thermal stability. The phase is readily
coarsened and softened at the temperatures above 393– 403 K.8,9) Additionally, the phase has a cubic crystal
structure that is incoherent with the h.c.p. magnesium matrix, leading to the fragility of the Mg/ interface. Al atoms diffuse quickly in the magnesium matrix, and then reduce the solid solution strengthening. All of the above phenomena result in poor creep properties of the Mg-8Al alloy, as presented in Fig. 8 and Table 3.
The RE-containing alloys exhibit good creep resistance due to have the Al11RE3intermetallic compounds. Although
Al11RE3 has a body-centered orthorhombic structure,22)
which is incoherent with the h.c.p. lattice of the magnesium
matrix, its high melting point (>1513K)23) and strong intermetallic bonding as well as the low diffusion rate of the RE elements in magnesium all contribute to a thermally stable alloy, even at 773 K.25,30)
3.3.2 Creep mechanism
Generally, the steady-state creep rate (""_s) can be
repre-sented by the relationship between stress () and temperature (T) as follows:
_
"
"s¼Anexp
Qc
RT
where A is a constant; n, the stress exponent; and Qc,
the apparent activation energy for creep. Figure 9 illustrates the stress dependence on steady-state strain rate (""_s).
Figure 10 shows an Arrhenius plot for the creep test of both Mg-8Al and Mg-8Al-2RE alloys under 40 MPa. The creep
0 5 10 15 20 25 30 35 40
Creep rupture life, t/h
0 10 20 30 40 5 15 25 35 Strain, ε (%) Mg-8Al-xRE alloy Creep test at 473 K
Mg-8Al Mg-8Al-1RE Mg-8Al-2RE Mg-8Al-3RE
(a)
0 100 200 300 400 500 600 700 800
Creep rupture life, t/h
0 20 40 60 10 30 50 Strain, ε (%) Mg-8Al-xRE alloy Creep test at 423 K
Mg-8Al Mg-8Al-1RE Mg-8Al-2RE Mg-8Al-3RE
(b)
Fig. 8 The typical constant-load creep curves of the alloys containing a series of RE tested at (a) 473 K and (b) 423 K, respectively, under 40 MPa.
100 90 80 70 60 50 40 30
Stress, σ/MPa
1e-008 1e-007 1e-006 1e-005 Strain rate , s /s -1
Mg-8Al, 473K n=2.0
Mg-8Al-2RE, 473K n=1.9
Mg-8Al, 423K n=2.1
Mg-8Al-2RE, 423K n=2.1
ε
Fig. 9 A plot showing the relationships of steady-state creep rate and applied stress at 423 and 473 K, respectively, for Mg-8Al-xRE alloys.
2.1 2.15 2.2 2.25 2.3 2.35 2.4
1000/T /K-1
1e-008 1e-007 1e-006 1e-005 Strain rate, ε s /s -1
Mg-8Al, Q=114 KJ mol-1
[image:6.595.312.544.70.280.2]Mg-8Al-2RE, Q=104 KJ mol-1
[image:6.595.48.289.72.572.2] [image:6.595.317.540.331.537.2]activation energy of the Mg-8Al and Mg-8Al-2RE alloys are 114 and 104 kJ mol1, respectively. For the Mg-8Al-xRE
alloys, n value was approximately 2, which suggests the creep of the alloys can be controlled by grain boundary sliding. This has previously been suggested for Mg-based alloys under similar test conditions.31–33)
3.3.3 Microstructural observations of the creep test specimens
The near grain boundary microstructure of post-crept Mg-8Al was shown in Fig. 11(a). Compared to the initial state (see Fig. 1(c)), the microstructure underwent significant changes during creep test. Fig. 11(a) shows that a high volume fraction of secondary discontinuous precipitates near grain boundaries during creep and many cracks are developed. Since the discontinuous precipitation occurring during creep effectively multiplies the grain boundary area available for easy deformation by grain boundary sliding in elevated temperature creep.31,32)Therefore, the creep
resist-ance of Mg-8Al should be poor at elevated temperatures. Figure 11(b) shows the microstructure near grain boun-dary of Mg-8Al-3RE after creep test. No significant change in microstructure morphology after creep exposure was observed. The addition of RE to Mg-8Al alloy forms the Al11RE3 intermetallic phase, which may suppress the
precipitation of discontinuousphase during the creep test. Consequently, the sliding of grain boundaries and the slip of dislocations in the matrix were effectively prevented at elevated temperatures, improving the creep resistance of Mg-8Al base alloy.34,35) Thus the alloying with RE leads to
increasing both the corrosion and creep resistance of Mg-8Al base alloy.
4. Conclusions
(1) The(Mg17Al12) and the Al11RE3phases are the main
precipitates in the Mg-8Al-xRE magnesium alloy matrix by the addition of the RE (La-rich Mischmetal). (2) RE additions suppress the precipitates of the thermally less stable phase at high temperatures and also stimulate the formation of the stable intermetallic-Al11RE3 phase at higher temperatures. The phase
becomes coarsening or even softening in the matrix,
and the strengthening effect of the phase gradually gets lost. However, at elevated temperature, Al11RE3
particles may still keep a role to obstruct both the dislocation motion and grain boundary sliding; hence, the creep property of the magnesium alloys may be still maintained.
(3) The stress exponents of the Mg-8Al-xRE alloys are approximately 2, which suggested creep mechanism of alloys was controlled by the grain boundary sliding. The creep activation energy of the 8Al and Mg-8Al-2RE alloys are 114 and 104 kJ mol1, respectively. (4) The corrosion tests indicate that the corrosion potential of these alloys becomes high and corrosion current density is reduced, as the RE content of the Mg-8Al alloys increases. The corrosion test of immersion also reveals that the rate of weight loss of the Mg-8Al alloy reduces with the RE content of the alloys.
Acknowledgement
The authors would like to appreciate the National Science Council of the Republic of China for financially supporting this research under Contract No. NSC93-2216-E-002-014.
REFERENCES
1) G. S. Cole: Mater. Sci. Forum419–422(2003) 43–49.
2) S. Schumann and H. Friedrich: Mater. Sci. Forum419–422(2003) 51– 56.
3) T. Kaneko and M. Suzuki: Mater. Sci. Forum419–422(2003) 67–72. 4) G. S. Cole and A. M. Sherman: Mater. Charact.35(1995) 3–9. 5) A. Stalmann, W. Sebastian, H. Friedrich, S. Schumann and K. Droder:
Adv. Eng. Mater.3(2001) 969–974. 6) A. A. Luo: JOM54(2002) 42–48.
7) F. Czerwinski and A. Zielinska-Lipiec: Acta Mater.53(2005) 3433– 3444.
8) K. S. Nair and M. C. Mittal: Mater. Sci. Forum30(1998) 89–93. 9) G. Pettersen, H. Westengen, R. Hoier and O. Lohne: Mater. Sci. Eng.
A207(1996) 115–120.
10) G. Yuan, Q. Wang and W. Ding: J. Mater. Sci.37(2002) 127–132. 11) Z. Zhang, R. Tremblay and D. Dube: Mater. Sci. Eng.A385(2004)
286–291.
12) Y. Mori, Y. Terada and T. Sato: Mater. Trans.46(2005) 1749–1752. 13) X. Gao, S. M. Zhu, B. C. Muddle and J. F. Nie: Scripta Mater.53(2005)
1321–1326.
(a)
10
µ
m
(b)
10
µ
m
[image:7.595.98.501.72.216.2]crack
14) M. Vogel, O. Kraft and E. Arzt: Metall. Mater. Trans. A36A(2005) 1713–1719.
15) J. Bai, Y. Sun, S. Xun, F. Xue and T. Zhu: Mater. Sci. Eng.A419
(2006) 181–188.
16) ASM Specialty Handbook, Magnesium and Magnesium Alloys, Amer-ican Society for Metals, (Materials Park, OH, 1999) pp. 14–15. 17) G. Yuan, M. Liu, W. Ding and I. Akihisa: Mater. Sci. Forum419–422
(2003) 425–431.
18) W. G. Yang and C. H. Koo: Mater. Trans.44(2003) 1029–1035. 19) W. G. Yang and C. H. Koo: Mater. Trans.44(2003) 1198–1203. 20) R. E. Reed-Hill: Physical Metallurgy Principles (PWS Publishing
Company Press, Boston 1994) p. 465.
21) L. Y. Wei and G. L. Dunlop: J. Alloys Compd.232(1996) 264–268. 22) P. Villars and L. O. Calvert (eds.):Pearson’s Handbook of
Crystallo-graphic Data for Intermetallic phases, American Society for Metals (Metals Park, OH, 1985) p. 1016.
23) Y. Li and H. Jones: Mater. Sci. Technol.12(1996) 651–661. 24) L. Y. Wei, G. L. Dunlop and H. Westengen: J. Mater. Sci.31(1996)
387–397.
25) L. Y. Wei, G. L. Dunlop and H. Westengen: Mater. Sci. Technol.12
(1996) 741–750.
26) J. I. Skar and D. Albright: Magnesium Technology (2002) pp. 255–261. 27) E. W. Frazier, E. W. Lee, M. E. Donnellan and J. J. Thompson: J. of
Metals May (1989) 22–26.
28) O. Lunder and K. Nisancioglu, in: Progress in the understanding and prevention of corrosion, The 10th European Corrosion Congress, 1993, The Institute of Material, London, (1993) p. 1249.
29) Y. Lu¨, Q. Wang, X. Zeng, W. Ding, C. Zhai and Y. Zhu: Mater. Sci. Eng.A278(2000) 66–76.
30) A. Luo and M. O. Pekguleryuz: J. Mater. Sci.29(1994) 5259–5271. 31) M. S. Dargusch, G. L. Dunlop and K. Pettersen: Proc of Conf. on
Magnesium Alloys and their Applications, B. L. Mordike and K. U. Kainer, Eds., Verkstoff-Informationsgesellschaft, Wolsburg, Germany, (1998) pp. 277–282.
32) H. Watanabe, T. Mukai, M. Kohzu, S. Tanabe and K. Higashi: Acta Mater.47(1999) 3753–3758.
33) C. J. Bettles: Mater. Sci. Eng.A348(2003) 280–288.
34) I. P. Moreno, T. K. Nandy, J. W. Jones, J. E. Allison and T. M. Pollock: Scripta Mater.48(2003) 1029–1034.
35) W. Du, Y. Sun, X. Min, F. Xue, M. Zhu and D. Wu: Mater. Sci. Eng.