The effect of grain size on the tensile and creep properties of Mg–2.6Nd–0.35Zn–xZr alloys at 250 °C
Introduction
MgNdZnZr based alloys in the composition ranges of (2.0–2.8)% Nd, (0.2–0.7)% Zn and (0.4–1.0)% Zr (in wt%) are commercially important creep resistant Mg alloys [1]. Compared with the more creep resistant yttrium-containing (the WE series) magnesium alloys, they are more cost affordable for applications up to 250 °C. The alloy system was first developed in Russia in the early 1960s for aerospace applications, known as ML10 [2], [3]. It was subsequently introduced into China and classified as ZM6, where ZM refers to cast Mg alloys and the digit 6 represents one of the 10 series sand cast Mg alloys in the category of aeronautical materials [1]. Over the last decade researchers have continued to develop magnesium-rare earth (RE) based sand cast creep-resistant Mg alloys for potential automotive and aerospace applications [4], [5], [6]. Our assessment is that the MgNdZn system will continue to be a prime candidate Mg system for long-term applications up to 250 °C. A recent detailed CALPHAD assessment of the MgNdZn system [7] has provided a useful thermodynamic basis for its further improvement.
The main role of each alloying element in the ML10 or ZM6 alloys can be summarized below. Nd is the prime strengthening element which forms fine Mg12(Zn,Nd) precipitates with Mg and Zn in the T6 microstructure [2], [3], [8]. The small presence of Zn (0.2%–0.7%) promotes precipitation of basal precipitate plates in the alloy in addition to prismatic precipitate plates [8]. Zr imparts to these alloys structural uniformity and consistency in performance through potent grain refinement [9].
The room temperature and elevated temperature (up to 300 °C) properties, including creep properties, of the ML10 or ZM6 alloys have been well documented in Ref. [1] based on detailed Russian and Chinese work. However, there is little systematic information about the effect of the grain size on the tensile and creep properties of these alloys at elevated temperatures. Commercial practice generally aims to produce a fine as-cast grain structure through a significant addition of Zr [10]. Bettles et al. [11], [12] have recently studied the influence of grain size (46–536 μm) on the tensile, compressive, creep and bolt load retention (BLR) properties of a similar Mg alloy (AM-SC1) in the T6 state up to 177 °C (similar to ML10 or ZM6, AM-SC1 typically contains 1.7% Nd, 1.0% other RE elements, 0.5% Zn, 1.0% Zr and balance Mg [12]). The yield strength of the alloy showed a clear Hall–Petch relationship at 177 °C with grain size in the range of 235–536 μm and the steady-state creep rate peaked at the grain size of 235 μm [12]. However, grain refinement below 200 μm did not show any significant effect on the properties of the AM-SC1 alloy at either 150 °C or 177 °C [12]. These results indicate that the influence of the grain size on the mechanical behavior of Mg–RE based alloys is not straightforward.
The purposes of this study are threefold: to assess the effect of grain size on the tensile and creep properties of Mg–2.65Nd–0.30Zn–xZr alloys at 250 °C; to identify the maximum T6 grain size, rather than the as-cast grain size, that can still meet the design performance criteria of the alloy at 250 °C; and to evaluate the effect of the solid solution treatment on grain growth and morphology. The motivation for the last two purposes is that commercial practice has long focused on achieving a fine equiaxed as-cast grain structure using excessive Zr. It is therefore of practical importance to clarify both issues.
Section snippets
Experimental procedure
Melting and alloying were conducted in a mild steel crucible placed in an electric resistance furnace. A melt with a nominal composition of Mg–2.65Nd–0.30Zn was prepared from pure Mg (99.8%), Zn (99.9%), and a Mg-30% Nd master alloy at 720 °C. The liquidus temperature of the alloy is 640 °C [1]. The melt was covered with a RJ2 fluxing agent (38–46% MgCl2, 32–40% KCl, 5–8%BaCl2, 3–5% CaF2, and 8% (NaCl+CaCl2)) and refined with a RJ5 fluxing agent (24–30% MgCl2, 20–26% KCl, 28–31% BaCl2, 13–15% CaF2
As-cast grain structures
Although grain refinement of ML10 or ZM6 alloys with Zr is established commercial practice, little information is available about their grain-refining characteristics as a function of the soluble Zr content. Fig. 1 shows the grain refinement results obtained. The representative grain structures are shown in Fig. 2. The following observations are notable.
- •
The Mg–2.6Nd–0.35Zn alloy without Zr solidified as largely columnar grains in the size range of 900±430 μm. This supports Emley's comment that
Equiaxed grain formation in the Mg–2.6Nd–0.35Zn alloy
A uniform equiaxed grain structure is important because it offers reliability and consistency in performance. The dominant columnar grains in the Mg–2.6Nd–0.35Zn alloy without Zr (Fig. 2(a), 900±430 μm) imply that the growth restriction effect from the 2.6% Nd and 0.35% Zn is negligible (under similar casting conditions Mg9Al1Zn will solidify as nearly equiaxed grains of ∼140 μm). A recent theoretical study [18] has confirmed that without considering the secondary effect such as dendrite
Summary
- •
Sand cast Mg–2.6Nd–0.35Zn alloy without grain refinement by Zr solidifies as largely columnar grains in the size range of 900±430 μm. An addition of 1.8% Zr (dissolved Zr: 0.41%) reduces the grain size from 900 μm to 84 μm but most grains are still irregular rather than equiaxed. The grain size shows a linear dependency on 1/Q calculated from the soluble Zr content. This, together with the columnar grains of the Mg–2.6Nd–0.35Zn alloy, suggests a negligible growth restriction effect of the presence
Acknowledgments
This work was supported by the Ministry of Science and Technology China through the Grant 2009GJB20011.
References (24)
- et al.
J. Alloys Compd.
(2011) - et al.
Mater. Sci. Eng. A
(2009) - et al.
Scr. Mater.
(2006) Acta Mater.
(2006)- et al.
Acta Mater.
(2010) - et al.
Mater. Sci. Eng. A
(2008) - et al.
Scr. Mater.
(2003) - et al.
Mater. Sci. Eng. A
(2006) - et al.
- et al.
Met. Sci. Heat Treat.
(1968)