- Email: [email protected]
Effect of addition of mutually soluble and insoluble metallic elements on the microstructure, tensile and compressive properties of pure magnesium
Materials Science and Engineering A 530 (2011) 149–160
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of addition of mutually soluble and insoluble metallic elements on the microstructure, tensile and compressive properties of pure magnesium S. Sankaranarayanan, S. Jayalakshmi, M. Gupta ∗ Department of Mechanical Engineering, National University of Singapore (NUS), 9 Engineering Drive 1, Singapore 117576, Singapore
a r t i c l e
i n f o
Article history: Received 18 July 2011 Received in revised form 16 September 2011 Accepted 17 September 2011 Available online 21 September 2011 Keywords: Magnesium Composites Metallic additions Mechanical alloying Casting Microstructure Mechanical properties Scanning electron microscopy (SEM)
a b s t r a c t In the present study, pure magnesium incorporated with metallic elements was synthesized using the disintegrated melt deposition technique followed by hot extrusion. The metallic elements added include: (i) mutually soluble element, aluminium (Al), (ii) insoluble element, titanium (Ti) and (iii) a combination of mutually soluble and insoluble elements (Al and Ti). The addition of the combination of elements was carried out into two ways: (a) addition after prior ball milling and (b) direct addition. The developed Mgbased materials were investigated for their microstructural and mechanical properties. Microstructural investigation revealed significant grain refinement due to metallic addition. The evaluation of mechanical properties showed significant improvement in microhardness, tensile and compressive properties of all the Mg-materials when compared to pure magnesium. The addition of individual elements resulted in the formation of Mg–3Al alloy and Mg–5.6Ti composite, and improved both the strength and ductility. When the elements were ball milled, Al3 Ti intermetallic was formed due to solid state reaction resulting in Mg–(3Al + 5.6Ti)BM composite, which was absent during direct addition (Mg–3Al–5.6Ti). The Mg–(3Al + 5.6Ti)BM composite showed the highest strength, however at the expense of ductility, while the Mg–3Al–5.6Ti showed relatively lower strength properties. The observed difference in behaviour between Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM is primarily attributed to the Al3 Ti intermetallic phase formation and the change in morphology and distribution of the metallic elements due to the ball-milling process. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Magnesium, the lightest of the structural metals with a density of ∼1740 kg/m3 , is ∼36% lighter than aluminium and ∼78% lighter than steel [1,2]. Pure Mg exhibits nominal specific strength along with excellent damping capacity and machinability. However, the applications of pure magnesium are rather limited because of its low strength, ductility and lack of high temperature stability [1,3]. Most of these limitations can be circumvented by the addition of selective alloying elements such as Al, Zn Zr, and RE metals. While Mg–alloys based on Al additions (such as AZ series, AM series and AS series) are the most commonly used alloys in the transportation sector, the other class of Mg–alloys (ZK series, ZC series, etc.) exhibits excellent creep resistance and are used in critical applications [4–7]. In spite of the development of new alloys, most of the commercially available Mg–alloys exhibit low thermal stability and are stable only up to ∼423 K [6,7]. While the incorporation
∗ Corresponding author. Tel.: +65 6516 6358; fax: +65 6 779 1459. E-mail address: [email protected] (M. Gupta). 0921-5093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2011.09.066
of ceramic reinforcements in pure Mg and its alloys (to form their metal matrix composites, Mg–MMCs) improves their strength, wear resistance and high temperature stability, they suffer from extreme brittleness [8–11]. Recently, studies on the addition of stronger and stiffer metallic elements like Ni, Cu, and Ti, to pure Mg/alloys have been conducted in order to understand their effect on the strength and ductility [12–15]. The aim of the current work is to improve the mechanical properties of pure Mg through metallic additions. The Mg-based materials are prepared by liquid-state processing using the disintegrated melt deposition (DMD) technique [10]. The metallic additions used are: (i) Al (mutually soluble in Mg), (ii) Ti (insoluble or negligibly soluble in Mg) and (iii) Al and Ti (a combination of soluble and insoluble elements), which are added to Mg with and without initial solid-state processing. A comparison is also made when these elements (Al and Ti) are added (a) directly and (b) after ball milling into pure Mg. Investigations are undertaken to study the effect of the addition of mutually soluble, insoluble metallic elements and their combination on the microstructural and mechanical properties, using the processing–structure–property relationship.
150
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
2. Experimental procedure
2.3. Materials characterization
2.1. Matrix material
2.3.1. Density measurements The experimental mass densities of pure Mg and the Mg-based materials developed were determined using Archimedes’ principle. This involved weighing the sample in air and then in distilled water using an A&D ER-182A electronic balance with an accuracy of ±10−7 kg. The theoretical densities of composites were calculated using the rule of mixtures [10].
Mg turnings of >99.9% purity (ACROS Organics, USA), was used as the matrix material. 2.1.1. Elemental metallic additions The metallic additions were: (i) addition of 3 wt.% Al particulates (based on the Al content in commercially available AZ31 alloy), which is mutually soluble in Mg, (ii) addition of 5.6 wt.% Ti particulates (based on our earlier work [15]), which is insoluble in Mg, (iii) addition of a combination of 3 wt.% Al and 5.6 wt.% Ti by (a) incorporation after pre-processing and (b) direct addition of 3 wt.% Al and 5.6 wt.% Ti (without any pre-processing). Elemental Ti particulates of particle size <140 × 10−6 m (purity 98%) supplied by Merck, and Al particulates of average particle size ∼15 × 10−6 m (purity 98%) supplied by Alfa Aesar were used. 2.1.2. Pre-processing of metallic additions The pre-processing of the metallic additions is carried out using ball-milling, which is a solid-state processing technique. Retsch PM-400 mechanical alloying machine was used for ball-milling the metallic elements (Al and Ti), herein referred to as (3Al + 5.6Ti)BM . Prior to ball milling, the elemental Ti and Al particulates were blended for 1 h (with 0.3 wt.% stearic acid as process control agent) so as to ensure uniform mixing of Ti and Al particulates. After blending, hardened stainless steel balls of diameter 0.015 m were added and the blended mixture was ball-milled for 2 h. Ball-milling was carried out to reduce the size of Al and Ti particulates and to allow the metallic particulates to react with each other [16]. The ball to powder weight ratio was kept at 20:1 and the speed of the milling machine was set at 200 rpm during the blending and ball milling processes [17]. 2.2. Processing 2.2.1. Melting and casting The Mg-based materials were prepared using the DMD technique [12,13]. The Mg turnings together with the metallic additions were heated in a graphite crucible to 1023 K in an electrical resistance furnace, under inert argon gas protective atmosphere. The superheated molten slurry was stirred for 5 min at 460 rpm using a twin blade (pitch 45◦ ) mild steel impeller. This facilitates uniform distribution of the incorporated metallic elements in the Mg-matrix. The impeller was coated with Zirtex 25 (86%ZrO2 , 8.8%Y2 O3 , 3.6%SiO2 , 1.2%K2 O and Na2 O, and 0.3% trace inorganic) to avoid iron contamination of the molten metal. The melt was then released through a 0.01 m diameter orifice at the base of the crucible and it was disintegrated by two jets of argon gas (at a flow rate of ∼25 lpm), which was located at a distance 0.265 m from the melt pouring point and oriented normal to the melt stream. The disintegrated melt slurry was subsequently deposited onto a metallic substrate located 0.5 m from the disintegration point. An ingot of 0.04 m diameter was obtained following the deposition stage. The synthesis of pure Mg was carried out in a similar fashion, except that no alloying elements were added. 2.2.2. Hot extrusion The pure Mg and the Mg-based materials obtained were machined to a diameter of 0.036 m and soaked at 673 K for 1 h. Hot extrusion was then carried out using a 150 T hydraulic press at 623 K with an extrusion ratio of 20.25:1 to obtain rods of 0.008 m in diameter. Samples from the extruded rods were used for further characterization as detailed in the next section.
2.3.2. Coefficient of thermal expansion An INSEIS TMA PT 1000LT thermo-mechanical analysis instrument was used to determine the coefficient of thermal expansion (CTE) of the developed Mg-based materials. Heating rate of 5 K/min was maintained. Argon gas flow rate was maintained at 0.1 lpm. Displacement of the test samples (each 0.005 m long) as a function of temperature (323–673 K) was measured using an alumina probe and was subsequently used to determine the CTE. 2.3.3. Microstructural characterization The grain morphology and the distribution of second phases/particulates in Mg matrix were studied on the aspolished samples. A Hitachi S-4300 field emission scanning electron microscope (FESEM) equipped with energy dispersive X-ray spectroscopy (EDS), a Jeol JSM-5800 LV scanning electron microscope (SEM), an Olympus metallographic optical microscope and Scion image analysis software were used for this purpose. The microstructure of the as-received Ti, Al and ball milled (3Al + 5.6Ti)BM powders were also investigated using SEM for determining the average particle size and particle morphology. 2.3.4. X-ray diffraction studies X-ray diffraction (XRD) analysis was carried out on polished samples of the developed Mg-materials and the ball milled (3Al + 5.6Ti)BM powder using an automated Shimadzu LAB X XRD6000 diffractometer. The samples were exposed to Cu K␣ radiation ˚ at a scanning speed of 2◦ /min. The Bragg angles and ( = 1.54056 A) the values of interplanar spacing, d, obtained were matched with standard values for Mg, Ti, Al and other related phases. 2.3.5. Microhardness measurements The microhardness measurements were carried out on the aspolished samples of pure Mg and other Mg-based materials using Matsuzawa MXT 50 automatic digital Microhardness tester. Vickers indenter under a test load of 25 gf and a dwell time of 15 s was used to perform the micro hardness tests in accordance with the ASTM standard E3 84-99. The tests were conducted on three samples for each composition for 10–15 repeatable readings. 2.3.6. Tensile behaviour A fully automated servo-hydraulic mechanical testing machine, Model-MTS 810 was used to determine the tensile properties of the Mg-based materials, in accordance with ASTM test method E8M96. The crosshead speed was set at 4 × 10−7 m/s. Specimens with 0.005 m diameter and 0.25 m gauge length was used. Instron 2630 – 100 series clip-on type extensometer was used to measure the failure strain. For each composition, a minimum of 5 tests were conducted to obtain repeatable values. 2.3.7. Compressive behaviour The compressive properties of the developed Mg-based materials along with pure Mg were determined in accordance with ASTM test method E9-89a using MTS 810 testing machine with a crosshead speed set at 6.67 × 10−7 m/s. Compressive test specimens of 0.008 m diameter with length to diameter ratio, l/d ∼1,
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
151
Table 1 Results of density, porosity and CTE measurements. S. No.
Material
CTE (×10−6 /K)
Density measurements 3
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
3
Theoretical density (g/cm )
Experimental density (g/cm )
1.7400 1.7588 1.8019 1.8221 1.8221
1.7397 1.7587 1.8002 1.8213 1.8218
were used. For each composition, a minimum of 5 tests were conducted to obtain repeatable values. 2.3.8. Fracture behaviour The fracture surface analyses of all the Mg-materials, tested under tension and compression, were studied using Jeol JSM-5800 LV SEM and Hitachi S-4300 FESEM. 3. Results 3.1. Macrostructure Macrostructural defects such as blowholes, cracks and pores were not observed on the as-cast ingot and extruded rods. The surface of both as-cast ingot and extruded rods were smooth and free from any circumferential defects. 3.2. Density measurements The results of the density measurements conducted on the Mg-based materials are tabulated in Table 1. From the values of experimental and theoretical densities, it can be seen that neardense materials have been developed. The volumetric porosity calculated from the theoretical and experimental density values indicate that among the materials developed, Mg–5.6Ti composite shows the highest porosity level. However, the porosity levels are relatively very low and are found to be <0.1% in all the samples. 3.3. Coefficient of thermal expansion Table 1 shows the thermal expansion coefficients of pure Mg and the developed Mg based materials with Al and Ti additions.
± ± ± ± ±
0.0015 0.0008 0.0017 0.0019 0.0014
Porosity (%) 0.02 0.01 0.09 0.04 0.02
28.52 30.49 27.22 29.05 26.13
When compared to pure Mg, Mg–5.6Ti shows a slight reduction in CTE value, whereas an increase in the CTE values is observed in both Mg–3Al and Mg–3Al–5.6Ti (after direct addition of Al and Ti). Mg–(3Al + 5.6Ti)BM composite shows the lowest value of CTE among all the materials. 3.4. X-ray diffraction Fig. 1 shows the X-ray diffraction pattern of the developed Mgbased materials and also that of the ball milled (3Al + 5.6Ti)BM powder. In Mg–3Al, in addition to Mg peaks, peaks corresponding to Mg17 Al12 phase are observed. In Mg–5.6Ti, only peaks corresponding to pure Mg and Ti are observed indicating that no phase formation occurs between Mg and Ti. The XRD pattern of the ball milled powder (3Al + 5.6Ti)BM shows prominent peaks corresponding to Ti. In addition, peaks identifiable with Al3 Ti intermetallic phase are also observed. In Mg–(3Al + 5.6Ti)BM , the peaks identified are that of Mg, Ti and Al3 Ti intermetallic phase. On the other hand, in Mg–3Al–5.6Ti, no Al3 Ti peak is observed. Rather, peaks corresponding to Mg17 Al12 intermetallic phase can be seen in addition to the Mg and Ti peaks. However, in both the cases, the peaks of Mg17 Al12 and Al3 Ti intermetallic phases are of low intensity and are not prominently seen possibly due to their very low volume fraction. 3.5. Microstructure The microstructure and grain characteristics of pure Mg, Mg–3Al alloy, Mg–5.6Ti, Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM are shown in Fig. 2(a)–(e) and Table 2. From Table 2 and Fig. 2, it can be seen that all the Mg-based materials developed show lower grain sizes (grain refinement) when compared to pure Mg. It can also be observed that the grain size of all the Mg-materials containing Al (both without Ti and with Ti, and with pre-processing and direct addition) show relatively fine grain sizes (<10 × 10−6 m), while that containing Ti alone (i.e. Mg–5.6Ti) show higher average grain size (15 × 10−6 m) that is only slightly lower than pure Mg. Fig. 3(a)–(d) shows the distribution of second phase/particles in the Mg-matrix. The microstructure of Mg–3Al alloy (Fig. 3(a)) reveals the presence of Mg17 Al12 intermetallic phase (confirmed by XRD, Fig. 1) at the grain boundary, mostly at the triple points. The size of the Mg17 Al12 intermetallic phase is observed to be <5 × 10−6 m. From the microstructures in Fig. 3(b)–(d), Table 2 Results of microstructure studies.a
Fig. 1. X-ray diffraction patterns of developed Mg-based materials and ball milled powder.
S. No.
Material
Grain size (×10−6 m)
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
21 9 15 10 8
a
Based on approximately 100 grains.
± ± ± ± ±
3 3 4 5 2
Aspect ratio 1.6 1.6 1.6 1.6 1.6
± ± ± ± ±
0.4 0.4 0.4 0.5 0.4
152
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
Fig. 2. Optical micrographs showing the grain characteristics of: (a) pure Mg, (b) Mg–3Al, (c) Mg–5.6Ti, (d) Mg–3Al–5.6Ti and (e) Mg–(3Al + 5.6Ti)BM .
it can be seen that the Ti particulates are found clustered in Mg–5.6Ti and Mg–3Al–5.6Ti (shown by arrows in Fig. 3(b) and (c)). In Mg–(3Al + 5.6Ti)BM , several rod-like particles are additionally observed (Fig. 3(d)). Fig. 4(a)–(d) shows the size and distribution of the particles. The uneven particle size of the as-received Ti and Al particulates are evident from the microstructure and image analysis results
(Fig. 4(a), (b) and (d)), while relatively uniform-sized particles are seen in the ball milled (3Al + 5.6Ti)BM powder (Fig. 4(c) and (d)). Also the size of the particulates in the (3Al + 5.6Ti)BM powder is small compared to the as-received Ti particulates (Fig. 4(d)). From the micrographs it can also be seen that the as-received Ti particulates (Fig. 4(a)) contain more number of sharp edges whereas the ball milled (3Al + 5.6Ti)BM powder contains Ti particles with blunted corners and rounded edges (Fig. 4(c)).
Table 3 Results of microhardness measurements.
3.6. Microhardness
S. No.
Material
Microhardness (Hv)
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
48 65 71 78 93
± ± ± ± ±
1 2 2 3 3
Table 3 shows the results of the microhardness measurements. Pure Mg shows the lowest hardness values. When compared to pure Mg, the addition of Al and Ti, whether added individually or in combination, significantly increases the hardness values. It can be seen that when compared to pure Mg, the hardness values of Mg–(3Al + 5.6Ti)BM has almost increased by ∼95%
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
153
Fig. 3. SEM micrographs showing the distribution of second phases/particles in the Mg-matrix of developed Mg based materials.
Fig. 4. SEM micrographs of: (a) as received Ti, (b) as-received Al and (c) ball milled powder. (d) Shows the comparison of the variation of particle size with the number of particles in the as-received and ball milled powders.
154
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
3.8. Compressive properties Table 5 shows the room temperature compression test results of the developed Mg-based materials. The compressive properties of several ceramic reinforced Mg–MMCs available from literature [21–23] have also been listed. Pure Mg has the lowest strength and the addition of soluble Al, increases both the yield and ultimate compressive strengths without affecting the compressive failure strain. The addition of insoluble element, Ti, only slightly increases the compressive yield strength value, but increases the ultimate compressive strength to ∼30% when compared to pure Mg. When incorporated with the combination of soluble and insoluble elements, the Mg–(3Al + 5.6Ti)BM shows the highest compressive yield and ultimate strengths. In comparison, the Mg–3Al–5.6Ti, wherein Al and Ti are directly added, show improved compressive strength properties with respect to pure Mg and Mg–5.6Ti, but much lower than that of Mg–3Al and Mg–(3Al + 5.6Ti)BM . Except for Mg–3Al, all the other materials show a large reduction in compressive failure strain values. Fig. 5(b) shows the representative engineering stress–strain curves of the materials tested under compression. 3.9. Fractography
Fig. 5. Engineering stress–strain curves of the Mg-based materials (a) tensile and (b) compression loading conditions, respectively.
(i.e. doubled in value). The increase in hardness values occurs in the following order: Mg < Mg–3Al < Mg–5.6Ti < Mg–3Al–5.6Ti < Mg–(3Al + 5.6Ti)BM .
3.7. Tensile properties The tensile test results of the developed Mg-based materials are shown in Table 4. The table also shows the tensile properties of several Mg-composites with ceramic reinforcements available from literature [11,18–21]. From the results, it can be observed that when compared to pure Mg, significant improvement in both yield strength and ultimate tensile strength can be achieved by the addition of mutually soluble Al, insoluble Ti, and the combination of both. Among all the materials, Mg–(3Al + 5.6Ti)BM shows the highest strength values. When compared to pure Mg, while an improvement in ductility can also be observed when the elements are added individually (i.e. in Mg–3Al and Mg–5.6Ti), the increment in strength is accompanied by a reduction in ductility when incorporated with the combination of elements, i.e. in Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM . Fig. 5(a) shows the representative engineering stress–strain curves of the Mg-based materials tested under tension.
Fig. 6 shows the tensile fracture surfaces of all the test materials. In the case of pure Mg and Mg–3Al alloy (Fig. 6(a) and (b)), fracture occurs by cleavage mode interspersed with ductile features. In Mg–5.6Ti, although the fracture showed ductile features (Fig. 6(c)), particle debonding is also observed (Fig. 6(d)). In Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM mixed mode fracture occurred with dominant brittle features. While particle debonding can be observed in Mg–3Al–5.6Ti (Fig. 6(e)), in Mg–(3Al + 5.6Ti)BM cracking occurs at the matrix/particle interface and is seen to extend into the matrix (arrow in Fig. 6(f)). Fig. 7(a)–(e) shows the fracture surfaces of all the test materials after compressive loading. In all the test samples, fracture occurred at ∼45◦ angle with respect to the compression test axis. The fractographic evidences indicated that pure Mg exhibited dominant shear failure and showed more shear bands when compared to other Mgbased materials, which on the other hand exhibited rough fracture surfaces with mixed mode of shear and brittle features. 4. Discussion 4.1. Synthesis Synthesis of pure Mg and the Mg-based materials was successfully carried out by the DMD process followed by hot extrusion. The visual observation of as-cast and extruded samples indicates the absence of any macro-defects which clearly designates the suitability of processing parameters used in this study [10,24]. There was no reaction between the Mg melt slurries with the graphite crucible [10,13], which attributes to the inability of Mg to form stable carbides. Also the inert Ar gas atmosphere has prevented the oxidation of Mg melt. The characterization results clearly indicate the feasibility of the DMD process as a potential fabrication technique for Mg based materials [24]. 4.2. Coefficient of thermal expansion Based on the results of CTE (Table 1), with respect to pure Mg, the slight increase in the thermal expansion coefficients of Mg–3Al–5.6Ti and Mg–3Al is due to the higher CTE value of Mg17 Al12 (∼29.3 × 10−6 /K) [25] when compared to pure Mg (∼27.1 × 10−6 /K) [1]. On the other hand, the decrease in CTE observed in Mg–5.6Ti and Mg–(3Al + 5.6Ti)BM is attributed to the
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
155
Table 4 Results of room temperature tensile testing. S. No.
Material
0.2 Yield strength (MPa)
Ultimate tensile strength (MPa)
Failure strain (%)
1 2 3 4 5 6 7 8 9 10
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM Mg–15.4SiC [11] (DMD) Mg–1.1Al2 O3 [18] (DMD) Mg–(TiB2 + TiC) [19] (remelting and dilution) Mg–10Mg2 Si [20] (powder metallurgy) Mg–22 vol%Saffil [21] (melt infiltration)
125 ± 9 161 ± 15 158 ± 6 167 ± 5 194 ± 2 155 ± 1 209 ± 1 95 ± 2 225 –
169 ± 11 250 ± 20 226 ± 6 236 ± 5 265 ± 2 207 ± 9 242 ± 3 298 ± 2 – 163
6.2 ± 0.7 7.5 ± 1.1 8.0 ± 1.5 4.8 ± 0.5 4.8 ± 0.6 1.4 ± 0.1 3.5 ± 0.3 2.4 ± 0.4 1.0 1.0
lower CTE values of Ti (∼9.1 × 10−6 /K) [1] and Al3 Ti intermetallic phase (∼13 × 10−6 /K), respectively [26].
4.3. Microstructural characterization 4.3.1. Effect of the addition of soluble element (Al) Microstructural characterization of Mg–3Al (Table 2) shows that the addition of mutually soluble element (Al) refines the grain size. It should be noted that Al is a major alloying element in Mg due to its high solid solubility [2,3,6]. Hence, the addition of 3Al to Mg forms the Mg–3Al alloy. It is well-known that during the solidification of Mg–Al alloys, primary-Mg solid solution with fine grains would be nucleated, along with the formation of Mg17 Al12 eutectic network distributed along the grain boundaries [27]. Further, the hot deformation of the cast ingot is believed to result in the breakdown of the Mg17 Al12 network structure into fine particles [27]. In the present case, it is evident from Fig. 1 (XRD) that the Mg–3Al alloy is composed of (Mg + Mg17 Al12 ) and from Figs. 2(a) and 3(a) it can be seen that the extruded Mg–3Al alloy consists of fine grains (<10 × 10−6 m) with fine individual Mg17 Al12 particles of size < 5 × 10−6 m near the triple grain edges. The formation of Mg17 Al12 intermetallic phase at grain edges also prevents grain growth during recrystallization [28]. In addition, as the extrusion was performed at ∼623 K (which is >0.5Tm of the Mg metal), it results in recrystallization, thereby leading to the formation of nearly equiaxed grains (Table 2) [28].
4.3.2. Effect of the addition of insoluble element (Ti) When Mg is incorporated with Ti (Mg–5.6Ti), structural analyses indicate that no intermetallic phases are formed (Figs. 1 and 3(b)). Based on the Mg–Ti binary phase diagram, the solubility of Ti in Mg is negligible and it does not form any intermetallic phase with Mg [29]. However, the good wettability of Ti by molten Mg facilitates good interfacial bonding [30] and hence the addition of insoluble element Ti to Mg can be regarded as a metallic reinforcement, i.e. Mg–5.6Ti composite which is composed of (Mg + Ti).
From the microstructure of Mg–5.6Ti (Fig. 3(b)), it is seen that the distribution of Ti is fairly uniform. However, the Ti particles are irregular in shape (sharp-edged and unevenly sized particles, Fig. 4(a)) and due to the tendency of irregularly shaped particles to agglomerate [31,32], clustering of particles can be dominantly observed. The grain size analyses indicate that grain refinement is observed when compared to pure Mg, but not when compared to Mg–3Al alloy (Table 2). This can be attributed to the larger size of the Ti-particles. It has been reported by various authors that small-sized second phases/particles effectively inhibit grain growth during recrystallization and significantly refine the grain size, when compared to second phase/reinforcement with larger sizes [17,28]. Hence, in Mg–5.6Ti, the presence of larger Ti particles with an average size range of ∼20–40 × 10−6 m results in coarser grain size when compared to that in Mg–3Al. 4.3.3. Effect of combined addition of Al and Ti metallic elements 4.3.3.1. Thermodynamics of Al–Ti system. Considering the Al–Ti phase diagram [29], the predominant equilibrium intermetallic phases are Ti3 Al, Ti2 Al5 , AlTi, Al2 Ti and Al3 Ti, respectively. Thermodynamic estimations show that the free energy of formation of Al3 Ti is estimated to be −33 kJ/mol, and is the lowest when compared to AlTi and Ti3 Al, as Ti2 Al5 and Al2 Ti are considered as transitional phases [33]. 4.3.3.2. Microstructure of ball milled powder. In the present work, although the composition of the powders used for ball-milling lies in the compositional range of AlTi intermetallic phase formation (65%Ti–35%Al), based on the free energy estimates it is expected that Al3 Ti phase is likely to form during the solid-state reaction. Such a behaviour was reported by Yang et al. [33] in their work on Mg–Al–Ti during reactive sintering. According to Yang et al., due to the absence of chemical reactions between Mg and Ti, and based on the free energy considerations, the Al3 Ti intermetallic phase is preferentially formed during solid-state reactive sintering [33]. In the present case, the XRD analysis confirm the formation of Al3 Ti (Fig. 1) although the peaks were not prominent, and contains the
Table 5 Results of room temperature compression testing. S. No.
Material
0.2 Compressive yield strength (MPa)
Ultimate compressive strength (MPa)
Failure strain (%)
1 2 3 4 5 6 7 8 9 11 12
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti) Mg–(3Al + 5.6Ti)BM Mg [21] Mg–22 vol%Saffil [21] RZ5 [21] RZ5-22 vol%Saffil [21] AZ91 + 10vol%SiC + 3wt.%Si (squeeze casting) [22] WE54-13 vol%SiC (PM) [23]
74 ± 3 127 ± 2 85 ± 3 104 ± 1 139 ± 6 – – – – 160 230
273 ± 11 437 ± 5 360 ± 5 378 ± 13 431 ± 14 169 339 308 445 350 370
22.7 ± 4.9 21.7 ± 4.3 13.6 ± 1.2 12.6 ± 1.3 12.9 ± 1.6 11.9 4.6 16.7 5.2 9 14
156
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
Fig. 6. Fractographs of Mg-based materials after tensile loading showing: (a) dominant cleavage fracture in pure Mg, (b) ductile and cleavage fracture in Mg–3Al, (c) ductile features in Mg–5.6Ti, (d) particle debonding in Mg–5.6Ti, (e) mixed mode fracture and particle debonding in Mg–3Al–5.6Ti and (f) microvoid coalescence and crack formation at the vicinity of the particle/matrix interface in Mg–(3Al + 5.6Ti)BM .
peaks of Ti which is due to the high wt.% of Ti. Several rod-like particles (marked by arrows) observed in the powder after ball-milling (Fig. 3(d)) also indicates the formation of Al3 Ti. Further, microstructural investigation of the ball milled powder reveal the change in the morphology of elemental particulates of Al and Ti after the ball milling process (Fig. 4(c)), with the sharp corners of the as-received particulates eventually being rounded due to ball milling. In addition, the distribution of number of particulates vs. particulate size also indicates the changes induced as a result of ball milling, such as the reduction in the size of larger Ti particulates (Fig. 4(d)). 4.3.3.3. Microstructure of Mg–(3Al + 5.6Ti)BM . Upon the incorporation of (3Al + 5.6Ti)BM powder in Mg, the bulk Mg–(3Al + 5.6Ti)BM specimen reveals the presence of Al3 Ti, as observed from the phase identification (Fig. 1, XRD). This is due to the non-reactivity of Mg and Ti [29], and owing to the free energy of formation
of Al3 Ti [33]. It should also be noted that no Mg17 Al12 phase was detected. The formation of Al3 Ti is also supported by SEM micrographs wherein several blocky or rod-shaped particles of size ∼10–50 × 10−6 m are evident. From earlier reports, the Al3 Ti usually forms as blocky, rod-shaped or needle-shaped particles, with the size varying between submicrons to several tens of microns [34–37]. Hence, the Mg–(3Al + 5.6Ti)BM can be considered as an intermetallic reinforced Mg-composite, which is composed of Mg + (Ti + Al3 Ti). In comparison, the X-ray diffraction pattern of Mg–3Al–5.6Ti, wherein Al and Ti have been added directly during Mg melting, shows the diffraction peaks of Mg17 Al12 indicating the formation of the eutectic phase (Fig. 1). The estimated free energy of formation of Mg17 Al12 phase (liquid state) reported in literature is found to be −40 kJ/mol at 1023 K [37], which is more negative than that of Al3 Ti [33]. Hence the direct addition of Al and Ti to Mg during
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
157
Fig. 7. Fractographs of Mg-based materials under compression showing dominant shear failure.
liquid-state processing favours the formation of Mg17 Al12 rather than that of Al3 Ti. Therefore, the direct addition of Al and Ti into Mg can be regarded as Mg–3Al composite with Ti as metallic reinforcement, i.e. Mg + (Ti + Mg17 Al12 ). Based on these observations, it is evident that the pre-processing of metallic elements (solid-state reactive milling) significantly influences the microstructure of the developed Mg-based materials. Further, when compared to the rest of the Mg-based materials developed, the addition of ball-milled powder to Mg, (i.e. in Mg–(3Al + 5.6Ti)BM composite), has resulted in the reduction of average grain size of the composite, which is evident from the microstructure and the image analysis results (Table 2, Figs. 2(e) and 4(d)). As mentioned earlier, ball milling significantly reduces the average size and shape of the particles, and increases the number of small sized particles (Fig. 4(c)). Thus, during recrystallization, the increase in the number of sites for grain nucleation (large
number of small sized particles), effectively reduces the matrix grain size [18]. Further, it should be noted that the change in morphology of the (3Al + 5.6Ti)BM powders due to ball milling also leads to lower porosity levels in Mg–(3Al + 5.6Ti)BM composite when compared to Mg–5.6Ti and Mg–3Al–5.6Ti. The relatively higher porosity levels observed in Mg–5.6Ti and Mg–3Al–5.6Ti can be attributed to the formation of metal free zones at the sharp corners of Ti particulates, as a result of the inability of high viscosity particulate-molten alloy slurry to negotiate the sharp corners [38]; this has been avoided in Mg–(3Al + 5.6Ti)BM due to the round edged morphology obtained in the particles after ball milling. 4.4. Mechanical properties Literature study reveals that the addition of alloying elements and reinforcements to pure Mg has successfully improved the strength properties [5,9]. However the strengthening effect
158
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
depends on the properties of alloying elements/reinforcements such as (i) morphology and distribution, (ii) high hardness and strength properties, (iii) ability for grain nucleation, (iv) obstruction to dislocation movements, and (v) effective load transfer from the matrix [39,40]. 4.4.1. Microhardness The microhardness measurements shown in Table 3 indicate an increase in the mean hardness values of the Mg based materials when compared to pure Mg. This increase in hardness values corresponds to the presence of comparatively harder second phases such as Ti in Mg–5.6Ti, Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM , Mg17 Al12 in Mg–3Al and Mg–3Al–5.6Ti and Al3 Ti in Mg–(3Al + 5.6Ti)BM . Although Mg–3Al has finer grain size and that the Mg17 Al12 phase has higher hardness than Ti (Mg17 Al12 ∼2.2 GPa and Ti ∼0.97 GPa [41,42]), the hardness value of Mg–3Al alloy is slightly lower than that of Mg–5.6Ti. This can be attributed to the relatively higher volume fraction of the sharp-edged Ti particulates and their distribution, as sharp-edged particulates tend to increase the stress concentration at particle corners resulting in increased dislocation density, thereby contributing to higher hardness values [43]. Among all the materials, the maximum mean hardness value observed is in the Mg–(3Al + 5.6Ti)BM composite, which is due to the inherent high hardness of Al3 Ti (∼4 GPa) [44]. 4.4.2. Tensile and compressive properties In the current investigation, the results from room temperature tensile and compression tests indicate a significant improvement in the strength properties of Mg incorporated with metallic elements. The effect of these additions on the mechanical properties has been discussed in the following sections. 4.4.2.1. Mg–3Al alloy. In the current work, the improvements in both strength and ductility obtained due to the addition of Al (Mg–3Al alloy) in tensile loading and improvement in strength and minimal adverse effect on ductility under compressive loading condition can be seen in Tables 4 and 5 and Fig. 5. As mentioned earlier, Al is one of the major alloying elements known to improve the strength properties of Mg, and due to its high solubility in Mg, it forms the eutectic Mg17 Al12 second phase [3,27]. Based on the microstructural analysis, it is evident that the addition of Al refines the grain size significantly. Further, due to the hot extrusion process, the presence of small sized individual Mg17 Al12 particles occur along the grain boundary, mostly at triple grain edges, rather than as eutectic network. Hence, in the Mg–3Al alloy the improvement in strength properties can be attributed to the following strengthening mechanisms: (i) strengthening due to the grain size reduction, due to the formation of refined and equiaxed grains [45], (ii) solution strengthening effect [28] and (iii) formation of Mg17 Al12 phase at the grain edges that impede dislocation motion [27]. Further, the lower Al content, the reduction in grain size and the presence of small sized second phase particulates have resulted in the improvement in ductility. Tensile fracture surface morphology of Mg–3Al alloy confirms the dominant ductile fracture due to plastic deformation (Fig. 6(b)) when compared to the fracture surface of pure Mg, which shows fracture by cleavage (Fig. 6(a)). Compressive fracture surfaces of both pure Mg and Mg–3Al alloy shows failure by shear band formation (Fig. 7(a) and (b)). 4.4.2.2. Mg–5.6Ti composite. From Tables 4 and 5, the improvement in mechanical properties of Mg when incorporated with Ti particulates (Mg–5.6Ti composite) is evident under both tension and compression, though the values are lower than Mg–3Al. Also, the Mg–5.6Ti composite exhibits strength and ductility values
that are similar to or higher than those of ceramic reinforced Mgcomposites, a few of which are listed in Tables 4 and 5. These results indicate that the addition of Ti particles strengthen the composite by efficient load transfer. Based on the microstructural observations and the inherent properties of Ti particulates, the improvement in mechanical properties when compared to pure Mg can be ascribed to the following strengthening effects: (i) the increase in the dislocation density due to the thermal residual stresses, which arises from the mismatch in the coefficient of thermal expansion between Mg and Ti, contribute to the increase in the yield strength, (ii) strength improvement due to grain refinement, (iii) the morphology of the Ti-particles, wherein the sharp-edged particles contribute to higher dislocation density due to the increased stress concentration at the pointed corners [43], and (iv) higher hardness of Ti which eventually increases the load carrying capacity (ultimate strength) [39,40,45]. While the higher inherent tensile fracture strain of Ti [46] increases the tensile fracture strain, due to its low strain value under compression [47], the fracture strain of Mg–5.6Ti has reduced. In comparison, Mg with other metallic reinforcements like Ni or Cu reported earlier [12,13] show low tensile ductility, which is due to the chemical activity between Mg and Ni/Cu forming brittle intermetallics with Mg such as Mg2 Cu and Mg2 Ni. In the present case, the improved behaviour (both strength and ductility) of Mg–5.6Ti is not only due to the absence of formation of Mg-based intermetallic phases [29], but also due to the good wettability and structural compatibility of Ti and Mg (HCP structure) [30]. Tensile fracture surface of Mg–5.6Ti show mixed mode fracture (ductile + cleavage) modes (Fig. 6(c) and (d)). As the hard Ti particles efficiently aid in the load bearing capacity, they resist crack propagation and thereby increase the tensile ductility [14]. However, as Ti is insoluble in Mg (chemical inertness), the matrix–particle interface would be predominantly governed by mechanical bonding due to the good wettability between Ti and Mg, rather than chemical bonding [48]. Hence when the maximum load is reached, rather than particle fracture, debonding of Ti particulates would occur (Fig. 6(d)). Under compressive loads, while shear failure is the dominant mode of fracture (shear + brittle mode), the fracture surface appears relatively rough when compared to pure Mg (Fig. 7(a) and (c)). 4.4.2.3. Mg–(3Al + 5.6Ti)BM composite. Based on the tensile and compression test results (Tables 4 and 5), it has been observed that when a combination of soluble Al and insoluble Ti is incorporated into pure Mg after pre-processing, the resulting Mg–(3Al + 5.6Ti)BM composite shows enhanced strength properties, at the expense of ductility. When compared to pure Mg, ∼50% increase in yield and tensile strengths are evident, and under compression, the yield strength has been doubled (∼100% increase) with significant increase in ultimate strength as well. The strengthening in Mg–(3Al + 5.6Ti)BM composite occur due to the following reasons: (i) presence of hard Ti particulates [41], (ii) formation of hard Al3 Ti intermetallic phase [44], (iii) significant grain refinement and (iv) the increase in the dislocation density due to the thermal residual stresses, owing to the mismatch in the CTE between Mg and Al3 Ti [1,26]. These together contribute towards the increase in the strength properties [39,40]. Although intermetallic phases such as Al3 Ti tend to reduce the ductility, they are considered to be advantageous over conventional ceramic reinforcements [49]. It has been reported that the CTE values of intermetallics are closer to that of metals and are higher than that of ceramics. Thus they reduce the residual stresses at the particle/matrix interface relatively, thereby improving the ductility (i.e. ductility loss is relatively less when compared to ceramic reinforcements) [49]. This is evident from Tables 4 and 5, wherein a comparison made with Mg-composites
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
containing ceramic reinforcements show drastic reduction in ductility. Further, the ductility values of both Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM composites are similar, in spite of the presence of Al3 Ti intermetallic phase in Mg–(3Al + 5.6Ti)BM . This is primarily due to the change in the morphology of the particulates. It has been reported that for composites containing angular or sharpedged particles, the stress concentration would occur at the pointed corner of the particle, and that the presence of such particles in the composite reduces the ductility of the composite [43]. Eliminating such particle corners is reported to render uniform stress distribution and improves the ductility without decreasing the strengthening effect [43]. Hence, in spite of the presence of brittle Al3 Ti intermetallic phase in Mg–(3Al + 5.6Ti)BM , the presence of particulates with rounded edges in Mg–(3Al + 5.6Ti)BM composite would lead to reduced stress concentration and uniform stress distribution, thereby exhibiting better ductility. The tensile fracture surface morphology of Mg–(3Al + 5.6Ti)BM composite showed mixed mode fracture features. From Fig. 6(f), it can be seen that the interface bonding is good but there are also regions at the particle/matrix interface that contains microvoids that coalesce to form cracks that extend into the matrix. Chawla and Chawla [50] reported that in discontinuously reinforced metal matrix composites with strong interface, the matrix is significantly work hardened due to the presence of the hard reinforcements. As a result, the matrix would be placed under a very large degree of constraint with an inability for strain relaxation to take place by deformation. This eventually results in the onset of void nucleation and propagation at the particle/matrix interface, as is evident from the microvoids and cracks seen in the present case. In contrast, in the Mg–3Al–5.6Ti composite, particle debonding was dominant (Fig. 6(e)). Under compression, both composites show failure by shear fracture, but the surfaces appear very rough when compared to other Mg-materials (Fig. 7(d) and (e)). 5. Conclusions The following are the conclusions that could be drawn from the present study: (1) DMD technique can successfully synthesize magnesium alloys and composites. (2) The addition of soluble element (Al) in lower content formed the Mg–3Al alloy with Mg17 Al12 eutectic phase that refined the grain size and improved the hardness, tensile and compressive strength properties. (3) The addition of insoluble element (Ti) formed the Mg–5.6Ti composite. Based on the inherent properties of Ti such as hardness, ductility, structural compatibility with Mg and the good wettability of Ti and Mg, the Mg–5.6Ti composites exhibited enhancement in strength and ductility, especially under tensile loads. (4) The effects of the addition of the combination of soluble and insoluble metallic elements primarily depend on their method of addition. (5) While the pre-processing of Al and Ti by ball-milling initiated solid state reaction resulting in Ti + Al3 Ti intermetallic phase, the direct addition of elements to Mg resulted in Ti + Mg17 Al12 phase formation. Hence the Mg–(3Al + 5.6Ti)BM composite was composed of Mg + (Ti + Al3 Ti) while that of Mg–3Al–5.6Ti was made of Mg + (Ti + Mg17 Al12 ). (6) The Mg–(3Al + 5.6Ti)BM composite showed significant grain refinement, lower CTE and substantial improvement in hardness, tensile and compressive strengths, with a loss in ductility. These can be attributed to the inherent properties of Al3 Ti intermetallic phase and a change in morphology of the
159
particles. In comparison, the Mg–3Al–5.6Ti composite showed lower strength properties. (7) With pure Mg as the matrix, a combination of superior mechanical properties can be achieved by careful selection and incorporation of metallic elements and their method of addition. Acknowledgements This work has been supported in part by the Agency for Science, Technology and Research (ASTAR) grant 092 137 0015 (WBS# R265-000-321-305). One of the authors, Mr. S. Sankaranarayanan acknowledges the NUS research scholarship for supporting this research for his graduate program. References [1] C.J. Smithells, Metals Reference Book, 5th ed., Butterworth’s & Co. Ltd., London, 1976. [2] E.F. Emley, Principles of Magnesium Technology, Pergamon Press, Oxford, 1966. [3] M.M. Avedesian, H. Baker, ASM Specialty Handbook. Magnesium and Magnesium Alloys, ASM International, OH, 1999. [4] G. Pettersen, Mater. Sci. Eng. A 207 (1996) 115–120. [5] W. Du, Y. Sun, X. Min, F. Xue, M. Zhu, D. Wu, Mater. Sci. Eng. A 356 (2003) 1–7. [6] I.J. Polmear, Mater. Sci. Technol. 10 (1994) 1–16. [7] A. Luo, M.O. Perquleryuz, J. Mater. Sci. 29 (1994) 5259–5271. [8] C. Fritze, H. Berek, K.U. Kainer, S. Mielke, B. Wielage, in: B.L. Mordike, K.U. Kainer (Eds.), Magnesium Alloys and Their Applications, Werkstoff-Informationsgesellschaft, Frankfurt, Germany, 1998, pp. 635–640. [9] K.U. Kainer, Metal Matrix Composites Custom Made Materials for Automotive and Aerospace Engineering, Wiley, 2006. [10] M. Gupta, M.O. Lai, D. Sarvanaranganathan, J. Mater. Sci. 35 (2000) 2155–2165. [11] S. Ugandhar, M. Gupta, S.K. Sinha, Compos. Struct. 72 (2006) 266–272. [12] S.F. Hassan, M. Gupta, J. Alloys Compd. 335 (2002) L10–L15. [13] S.F. Hassan, M. Gupta, Mater. Res. Bull. 37 (2002) 377–389. [14] P. Peˇırez, G. Garceˇı s, P. Adeva, Compos. Sci. Technol. 64 (2004) 145–151. [15] S.F. Hassan, M. Gupta, J. Alloys Compd. 345 (2002) 246–251. [16] C. Suryanarayana, Mechanical Alloying and Milling, CRC Press, New York, 2004. [17] M.K. Habibi, S.P. Joshi, M. Gupta, Acta Mater. 58 (2010) 6104–6114. [18] S.F. Hassan, M. Gupta, J. Alloys Compd. 419 (2006) 84–90. [19] X. Zhang, H. Wang, L. Liao, X. Teng, N. Ma, Mater. Lett. 59 (2005) 2105–2109. [20] L. Lu, K.K. Thong, M. Gupta, Compos. Sci. Technol. 63 (2003) 627–632. [21] D.J. Towle, C.M. Friend, Mater. Sci. Technol. 9 (1993) 35–41. [22] Z. Trojanova, V. Gartnerova, A. Jager, A. Namesny, M. Chalupova, P. Palcek, P. Lukac, Compos. Sci. Technol. 69 (2009) 2256–2264. [23] Z. Szaraz, Z. Trojanova, M. Cabbibo, E. Evangelista, Mater. Sci. Eng. A 462 (2007) 225–229. [24] S. Sankaranarayanan, S. Jayalakshmi, M. Gupta, J. Alloys Compd. 509 (2011) 7229–7237. [25] V. Goryany, P.J. Mauk, J. Miner. Metall. 43 (2007) 85–97. [26] T. Li, E.A. Olevsky, M.A. Meyers, Mater. Sci. Eng. A 473 (2008) 49–57. [27] A. Kielbus, T. Rzychon, R. Chibis, J. Achiev. Mater. Manuf. Eng. 18 (2006) 135–138. [28] Z. Li., J. Dong., X.Q. Zeng., C. Lu., W.J. Ding., Mater. Sci. Eng. A 466 (2007) 134–139. [29] Binary Alloy Phase Diagram [Electronic Resource], 2nd ed., Plus Updates, ASM International, Materials Park, OH, 1996. [30] K. Kondoh, M. Kawakami, H. Imai, J. Umeda, H. Fujii, Acta Mater. 58 (2010) 606–614. [31] Z. Wang, M. Song, C. Sun, Y. He, Mater. Sci. Eng. A 528 (2011) 1131–1137. [32] A. Slipenyuk, V. Kuprin, Yu. Milmana, J.E. Spowart, D.B. Miracle, Mater. Sci. Eng. A 381 (2004) 165–170. [33] Z.R. Yang, S.Q. Wang, M.J. Gao, Y.T. Zhao, K.M. Chen, X.H. Cui, Composites A 39 (2008) 1427–1432. [34] X. Wang, A. Jha, R. Brydson, Mater. Sci. Eng. A 364 (2004) 339–345. [35] Y. Watanabe, Y. Fukui, Multidiscip. Microsc. Res. Edu. 18 (2004) 9–198. [36] Y.-h. Zhao, J. Zhou, W. Yan, Trans. Non-ferrous Met. Soc. China 12 (2002) 643–648. [37] D.W. Zhou, J.S. Liu, S.H. Xu, P. Peng, Physica B 405 (2010) 2863–2868. [38] M. Gupta, M.O. Lai, C.Y. Soo, Mater. Res. Bull. 30 (1995) 1525–1534. [39] D.J. Lloyd, Int. Mater. Rev. 39 (1994) 1–23. [40] I.A. Ibrahim, F.A. Mohamed, E.J. Lavernia, J. Mater. Sci. 26 (1991) 1137–1156. [41] ASM Handbook Properties and Selection: Non-Ferrous Alloys and SpecialPurpose Materials, vol. 2, ASM International, Materials, Park, OH, 1990. [42] M.S. Yoo, J.J. Kim, K.S. Shin, N.J. Kim, in: H.I. Kaplan (Ed.), Magnesium Technology, The Minerals, Metals & Materials Society, TMS, 2002. [43] S. Qin, C. Chen, G. Zhang, W. Wang, Z. Wang, Mater. Sci. Eng. A 272 (1999) 363–370. [44] Yu.V. Milmana, D.B. Miracleb, S.I. Chugunovaa, I.V. Voskoboinika, N.P. Korzhovaa, T.N. Legkayaa, Yu.N. Podrezova, Intermetallics 9 (2001) 839–845. [45] G.E. Dieter, Mechanical Metallurgy, McGraw-Hill, USA, 1986.
160
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
[46] Metals Hand Book – Properties and Selection of Metals, 8th ed., American Society for Metals, Metals Park, OH, p. 1225. [47] P.F. Barrett, Compressive Properties of Titanium Sheet at Elevated Temperatures, Technical Note 2038, National Advisory Committee for Aeronautics.
[48] S. Ahmed, F.R. Jones, Composites 21 (1990) 81–84. [49] Z.R. Yang, S.Q. Wang, X.H. Cui, Yu.T. Zhao, M.J. Gao, M.X. Wei, Sci. Technol. Adv. Mater. 9 (2008). [50] N. Chawla, K.K. Chawla, Metal Matrix Composites, Springer, New York, 2006.
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering A journal homepage: www.elsevier.com/locate/msea
Effect of addition of mutually soluble and insoluble metallic elements on the microstructure, tensile and compressive properties of pure magnesium S. Sankaranarayanan, S. Jayalakshmi, M. Gupta ∗ Department of Mechanical Engineering, National University of Singapore (NUS), 9 Engineering Drive 1, Singapore 117576, Singapore
a r t i c l e
i n f o
Article history: Received 18 July 2011 Received in revised form 16 September 2011 Accepted 17 September 2011 Available online 21 September 2011 Keywords: Magnesium Composites Metallic additions Mechanical alloying Casting Microstructure Mechanical properties Scanning electron microscopy (SEM)
a b s t r a c t In the present study, pure magnesium incorporated with metallic elements was synthesized using the disintegrated melt deposition technique followed by hot extrusion. The metallic elements added include: (i) mutually soluble element, aluminium (Al), (ii) insoluble element, titanium (Ti) and (iii) a combination of mutually soluble and insoluble elements (Al and Ti). The addition of the combination of elements was carried out into two ways: (a) addition after prior ball milling and (b) direct addition. The developed Mgbased materials were investigated for their microstructural and mechanical properties. Microstructural investigation revealed significant grain refinement due to metallic addition. The evaluation of mechanical properties showed significant improvement in microhardness, tensile and compressive properties of all the Mg-materials when compared to pure magnesium. The addition of individual elements resulted in the formation of Mg–3Al alloy and Mg–5.6Ti composite, and improved both the strength and ductility. When the elements were ball milled, Al3 Ti intermetallic was formed due to solid state reaction resulting in Mg–(3Al + 5.6Ti)BM composite, which was absent during direct addition (Mg–3Al–5.6Ti). The Mg–(3Al + 5.6Ti)BM composite showed the highest strength, however at the expense of ductility, while the Mg–3Al–5.6Ti showed relatively lower strength properties. The observed difference in behaviour between Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM is primarily attributed to the Al3 Ti intermetallic phase formation and the change in morphology and distribution of the metallic elements due to the ball-milling process. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Magnesium, the lightest of the structural metals with a density of ∼1740 kg/m3 , is ∼36% lighter than aluminium and ∼78% lighter than steel [1,2]. Pure Mg exhibits nominal specific strength along with excellent damping capacity and machinability. However, the applications of pure magnesium are rather limited because of its low strength, ductility and lack of high temperature stability [1,3]. Most of these limitations can be circumvented by the addition of selective alloying elements such as Al, Zn Zr, and RE metals. While Mg–alloys based on Al additions (such as AZ series, AM series and AS series) are the most commonly used alloys in the transportation sector, the other class of Mg–alloys (ZK series, ZC series, etc.) exhibits excellent creep resistance and are used in critical applications [4–7]. In spite of the development of new alloys, most of the commercially available Mg–alloys exhibit low thermal stability and are stable only up to ∼423 K [6,7]. While the incorporation
∗ Corresponding author. Tel.: +65 6516 6358; fax: +65 6 779 1459. E-mail address: [email protected] (M. Gupta). 0921-5093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2011.09.066
of ceramic reinforcements in pure Mg and its alloys (to form their metal matrix composites, Mg–MMCs) improves their strength, wear resistance and high temperature stability, they suffer from extreme brittleness [8–11]. Recently, studies on the addition of stronger and stiffer metallic elements like Ni, Cu, and Ti, to pure Mg/alloys have been conducted in order to understand their effect on the strength and ductility [12–15]. The aim of the current work is to improve the mechanical properties of pure Mg through metallic additions. The Mg-based materials are prepared by liquid-state processing using the disintegrated melt deposition (DMD) technique [10]. The metallic additions used are: (i) Al (mutually soluble in Mg), (ii) Ti (insoluble or negligibly soluble in Mg) and (iii) Al and Ti (a combination of soluble and insoluble elements), which are added to Mg with and without initial solid-state processing. A comparison is also made when these elements (Al and Ti) are added (a) directly and (b) after ball milling into pure Mg. Investigations are undertaken to study the effect of the addition of mutually soluble, insoluble metallic elements and their combination on the microstructural and mechanical properties, using the processing–structure–property relationship.
150
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
2. Experimental procedure
2.3. Materials characterization
2.1. Matrix material
2.3.1. Density measurements The experimental mass densities of pure Mg and the Mg-based materials developed were determined using Archimedes’ principle. This involved weighing the sample in air and then in distilled water using an A&D ER-182A electronic balance with an accuracy of ±10−7 kg. The theoretical densities of composites were calculated using the rule of mixtures [10].
Mg turnings of >99.9% purity (ACROS Organics, USA), was used as the matrix material. 2.1.1. Elemental metallic additions The metallic additions were: (i) addition of 3 wt.% Al particulates (based on the Al content in commercially available AZ31 alloy), which is mutually soluble in Mg, (ii) addition of 5.6 wt.% Ti particulates (based on our earlier work [15]), which is insoluble in Mg, (iii) addition of a combination of 3 wt.% Al and 5.6 wt.% Ti by (a) incorporation after pre-processing and (b) direct addition of 3 wt.% Al and 5.6 wt.% Ti (without any pre-processing). Elemental Ti particulates of particle size <140 × 10−6 m (purity 98%) supplied by Merck, and Al particulates of average particle size ∼15 × 10−6 m (purity 98%) supplied by Alfa Aesar were used. 2.1.2. Pre-processing of metallic additions The pre-processing of the metallic additions is carried out using ball-milling, which is a solid-state processing technique. Retsch PM-400 mechanical alloying machine was used for ball-milling the metallic elements (Al and Ti), herein referred to as (3Al + 5.6Ti)BM . Prior to ball milling, the elemental Ti and Al particulates were blended for 1 h (with 0.3 wt.% stearic acid as process control agent) so as to ensure uniform mixing of Ti and Al particulates. After blending, hardened stainless steel balls of diameter 0.015 m were added and the blended mixture was ball-milled for 2 h. Ball-milling was carried out to reduce the size of Al and Ti particulates and to allow the metallic particulates to react with each other [16]. The ball to powder weight ratio was kept at 20:1 and the speed of the milling machine was set at 200 rpm during the blending and ball milling processes [17]. 2.2. Processing 2.2.1. Melting and casting The Mg-based materials were prepared using the DMD technique [12,13]. The Mg turnings together with the metallic additions were heated in a graphite crucible to 1023 K in an electrical resistance furnace, under inert argon gas protective atmosphere. The superheated molten slurry was stirred for 5 min at 460 rpm using a twin blade (pitch 45◦ ) mild steel impeller. This facilitates uniform distribution of the incorporated metallic elements in the Mg-matrix. The impeller was coated with Zirtex 25 (86%ZrO2 , 8.8%Y2 O3 , 3.6%SiO2 , 1.2%K2 O and Na2 O, and 0.3% trace inorganic) to avoid iron contamination of the molten metal. The melt was then released through a 0.01 m diameter orifice at the base of the crucible and it was disintegrated by two jets of argon gas (at a flow rate of ∼25 lpm), which was located at a distance 0.265 m from the melt pouring point and oriented normal to the melt stream. The disintegrated melt slurry was subsequently deposited onto a metallic substrate located 0.5 m from the disintegration point. An ingot of 0.04 m diameter was obtained following the deposition stage. The synthesis of pure Mg was carried out in a similar fashion, except that no alloying elements were added. 2.2.2. Hot extrusion The pure Mg and the Mg-based materials obtained were machined to a diameter of 0.036 m and soaked at 673 K for 1 h. Hot extrusion was then carried out using a 150 T hydraulic press at 623 K with an extrusion ratio of 20.25:1 to obtain rods of 0.008 m in diameter. Samples from the extruded rods were used for further characterization as detailed in the next section.
2.3.2. Coefficient of thermal expansion An INSEIS TMA PT 1000LT thermo-mechanical analysis instrument was used to determine the coefficient of thermal expansion (CTE) of the developed Mg-based materials. Heating rate of 5 K/min was maintained. Argon gas flow rate was maintained at 0.1 lpm. Displacement of the test samples (each 0.005 m long) as a function of temperature (323–673 K) was measured using an alumina probe and was subsequently used to determine the CTE. 2.3.3. Microstructural characterization The grain morphology and the distribution of second phases/particulates in Mg matrix were studied on the aspolished samples. A Hitachi S-4300 field emission scanning electron microscope (FESEM) equipped with energy dispersive X-ray spectroscopy (EDS), a Jeol JSM-5800 LV scanning electron microscope (SEM), an Olympus metallographic optical microscope and Scion image analysis software were used for this purpose. The microstructure of the as-received Ti, Al and ball milled (3Al + 5.6Ti)BM powders were also investigated using SEM for determining the average particle size and particle morphology. 2.3.4. X-ray diffraction studies X-ray diffraction (XRD) analysis was carried out on polished samples of the developed Mg-materials and the ball milled (3Al + 5.6Ti)BM powder using an automated Shimadzu LAB X XRD6000 diffractometer. The samples were exposed to Cu K␣ radiation ˚ at a scanning speed of 2◦ /min. The Bragg angles and ( = 1.54056 A) the values of interplanar spacing, d, obtained were matched with standard values for Mg, Ti, Al and other related phases. 2.3.5. Microhardness measurements The microhardness measurements were carried out on the aspolished samples of pure Mg and other Mg-based materials using Matsuzawa MXT 50 automatic digital Microhardness tester. Vickers indenter under a test load of 25 gf and a dwell time of 15 s was used to perform the micro hardness tests in accordance with the ASTM standard E3 84-99. The tests were conducted on three samples for each composition for 10–15 repeatable readings. 2.3.6. Tensile behaviour A fully automated servo-hydraulic mechanical testing machine, Model-MTS 810 was used to determine the tensile properties of the Mg-based materials, in accordance with ASTM test method E8M96. The crosshead speed was set at 4 × 10−7 m/s. Specimens with 0.005 m diameter and 0.25 m gauge length was used. Instron 2630 – 100 series clip-on type extensometer was used to measure the failure strain. For each composition, a minimum of 5 tests were conducted to obtain repeatable values. 2.3.7. Compressive behaviour The compressive properties of the developed Mg-based materials along with pure Mg were determined in accordance with ASTM test method E9-89a using MTS 810 testing machine with a crosshead speed set at 6.67 × 10−7 m/s. Compressive test specimens of 0.008 m diameter with length to diameter ratio, l/d ∼1,
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
151
Table 1 Results of density, porosity and CTE measurements. S. No.
Material
CTE (×10−6 /K)
Density measurements 3
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
3
Theoretical density (g/cm )
Experimental density (g/cm )
1.7400 1.7588 1.8019 1.8221 1.8221
1.7397 1.7587 1.8002 1.8213 1.8218
were used. For each composition, a minimum of 5 tests were conducted to obtain repeatable values. 2.3.8. Fracture behaviour The fracture surface analyses of all the Mg-materials, tested under tension and compression, were studied using Jeol JSM-5800 LV SEM and Hitachi S-4300 FESEM. 3. Results 3.1. Macrostructure Macrostructural defects such as blowholes, cracks and pores were not observed on the as-cast ingot and extruded rods. The surface of both as-cast ingot and extruded rods were smooth and free from any circumferential defects. 3.2. Density measurements The results of the density measurements conducted on the Mg-based materials are tabulated in Table 1. From the values of experimental and theoretical densities, it can be seen that neardense materials have been developed. The volumetric porosity calculated from the theoretical and experimental density values indicate that among the materials developed, Mg–5.6Ti composite shows the highest porosity level. However, the porosity levels are relatively very low and are found to be <0.1% in all the samples. 3.3. Coefficient of thermal expansion Table 1 shows the thermal expansion coefficients of pure Mg and the developed Mg based materials with Al and Ti additions.
± ± ± ± ±
0.0015 0.0008 0.0017 0.0019 0.0014
Porosity (%) 0.02 0.01 0.09 0.04 0.02
28.52 30.49 27.22 29.05 26.13
When compared to pure Mg, Mg–5.6Ti shows a slight reduction in CTE value, whereas an increase in the CTE values is observed in both Mg–3Al and Mg–3Al–5.6Ti (after direct addition of Al and Ti). Mg–(3Al + 5.6Ti)BM composite shows the lowest value of CTE among all the materials. 3.4. X-ray diffraction Fig. 1 shows the X-ray diffraction pattern of the developed Mgbased materials and also that of the ball milled (3Al + 5.6Ti)BM powder. In Mg–3Al, in addition to Mg peaks, peaks corresponding to Mg17 Al12 phase are observed. In Mg–5.6Ti, only peaks corresponding to pure Mg and Ti are observed indicating that no phase formation occurs between Mg and Ti. The XRD pattern of the ball milled powder (3Al + 5.6Ti)BM shows prominent peaks corresponding to Ti. In addition, peaks identifiable with Al3 Ti intermetallic phase are also observed. In Mg–(3Al + 5.6Ti)BM , the peaks identified are that of Mg, Ti and Al3 Ti intermetallic phase. On the other hand, in Mg–3Al–5.6Ti, no Al3 Ti peak is observed. Rather, peaks corresponding to Mg17 Al12 intermetallic phase can be seen in addition to the Mg and Ti peaks. However, in both the cases, the peaks of Mg17 Al12 and Al3 Ti intermetallic phases are of low intensity and are not prominently seen possibly due to their very low volume fraction. 3.5. Microstructure The microstructure and grain characteristics of pure Mg, Mg–3Al alloy, Mg–5.6Ti, Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM are shown in Fig. 2(a)–(e) and Table 2. From Table 2 and Fig. 2, it can be seen that all the Mg-based materials developed show lower grain sizes (grain refinement) when compared to pure Mg. It can also be observed that the grain size of all the Mg-materials containing Al (both without Ti and with Ti, and with pre-processing and direct addition) show relatively fine grain sizes (<10 × 10−6 m), while that containing Ti alone (i.e. Mg–5.6Ti) show higher average grain size (15 × 10−6 m) that is only slightly lower than pure Mg. Fig. 3(a)–(d) shows the distribution of second phase/particles in the Mg-matrix. The microstructure of Mg–3Al alloy (Fig. 3(a)) reveals the presence of Mg17 Al12 intermetallic phase (confirmed by XRD, Fig. 1) at the grain boundary, mostly at the triple points. The size of the Mg17 Al12 intermetallic phase is observed to be <5 × 10−6 m. From the microstructures in Fig. 3(b)–(d), Table 2 Results of microstructure studies.a
Fig. 1. X-ray diffraction patterns of developed Mg-based materials and ball milled powder.
S. No.
Material
Grain size (×10−6 m)
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
21 9 15 10 8
a
Based on approximately 100 grains.
± ± ± ± ±
3 3 4 5 2
Aspect ratio 1.6 1.6 1.6 1.6 1.6
± ± ± ± ±
0.4 0.4 0.4 0.5 0.4
152
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
Fig. 2. Optical micrographs showing the grain characteristics of: (a) pure Mg, (b) Mg–3Al, (c) Mg–5.6Ti, (d) Mg–3Al–5.6Ti and (e) Mg–(3Al + 5.6Ti)BM .
it can be seen that the Ti particulates are found clustered in Mg–5.6Ti and Mg–3Al–5.6Ti (shown by arrows in Fig. 3(b) and (c)). In Mg–(3Al + 5.6Ti)BM , several rod-like particles are additionally observed (Fig. 3(d)). Fig. 4(a)–(d) shows the size and distribution of the particles. The uneven particle size of the as-received Ti and Al particulates are evident from the microstructure and image analysis results
(Fig. 4(a), (b) and (d)), while relatively uniform-sized particles are seen in the ball milled (3Al + 5.6Ti)BM powder (Fig. 4(c) and (d)). Also the size of the particulates in the (3Al + 5.6Ti)BM powder is small compared to the as-received Ti particulates (Fig. 4(d)). From the micrographs it can also be seen that the as-received Ti particulates (Fig. 4(a)) contain more number of sharp edges whereas the ball milled (3Al + 5.6Ti)BM powder contains Ti particles with blunted corners and rounded edges (Fig. 4(c)).
Table 3 Results of microhardness measurements.
3.6. Microhardness
S. No.
Material
Microhardness (Hv)
1 2 3 4 5
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM
48 65 71 78 93
± ± ± ± ±
1 2 2 3 3
Table 3 shows the results of the microhardness measurements. Pure Mg shows the lowest hardness values. When compared to pure Mg, the addition of Al and Ti, whether added individually or in combination, significantly increases the hardness values. It can be seen that when compared to pure Mg, the hardness values of Mg–(3Al + 5.6Ti)BM has almost increased by ∼95%
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
153
Fig. 3. SEM micrographs showing the distribution of second phases/particles in the Mg-matrix of developed Mg based materials.
Fig. 4. SEM micrographs of: (a) as received Ti, (b) as-received Al and (c) ball milled powder. (d) Shows the comparison of the variation of particle size with the number of particles in the as-received and ball milled powders.
154
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
3.8. Compressive properties Table 5 shows the room temperature compression test results of the developed Mg-based materials. The compressive properties of several ceramic reinforced Mg–MMCs available from literature [21–23] have also been listed. Pure Mg has the lowest strength and the addition of soluble Al, increases both the yield and ultimate compressive strengths without affecting the compressive failure strain. The addition of insoluble element, Ti, only slightly increases the compressive yield strength value, but increases the ultimate compressive strength to ∼30% when compared to pure Mg. When incorporated with the combination of soluble and insoluble elements, the Mg–(3Al + 5.6Ti)BM shows the highest compressive yield and ultimate strengths. In comparison, the Mg–3Al–5.6Ti, wherein Al and Ti are directly added, show improved compressive strength properties with respect to pure Mg and Mg–5.6Ti, but much lower than that of Mg–3Al and Mg–(3Al + 5.6Ti)BM . Except for Mg–3Al, all the other materials show a large reduction in compressive failure strain values. Fig. 5(b) shows the representative engineering stress–strain curves of the materials tested under compression. 3.9. Fractography
Fig. 5. Engineering stress–strain curves of the Mg-based materials (a) tensile and (b) compression loading conditions, respectively.
(i.e. doubled in value). The increase in hardness values occurs in the following order: Mg < Mg–3Al < Mg–5.6Ti < Mg–3Al–5.6Ti < Mg–(3Al + 5.6Ti)BM .
3.7. Tensile properties The tensile test results of the developed Mg-based materials are shown in Table 4. The table also shows the tensile properties of several Mg-composites with ceramic reinforcements available from literature [11,18–21]. From the results, it can be observed that when compared to pure Mg, significant improvement in both yield strength and ultimate tensile strength can be achieved by the addition of mutually soluble Al, insoluble Ti, and the combination of both. Among all the materials, Mg–(3Al + 5.6Ti)BM shows the highest strength values. When compared to pure Mg, while an improvement in ductility can also be observed when the elements are added individually (i.e. in Mg–3Al and Mg–5.6Ti), the increment in strength is accompanied by a reduction in ductility when incorporated with the combination of elements, i.e. in Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM . Fig. 5(a) shows the representative engineering stress–strain curves of the Mg-based materials tested under tension.
Fig. 6 shows the tensile fracture surfaces of all the test materials. In the case of pure Mg and Mg–3Al alloy (Fig. 6(a) and (b)), fracture occurs by cleavage mode interspersed with ductile features. In Mg–5.6Ti, although the fracture showed ductile features (Fig. 6(c)), particle debonding is also observed (Fig. 6(d)). In Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM mixed mode fracture occurred with dominant brittle features. While particle debonding can be observed in Mg–3Al–5.6Ti (Fig. 6(e)), in Mg–(3Al + 5.6Ti)BM cracking occurs at the matrix/particle interface and is seen to extend into the matrix (arrow in Fig. 6(f)). Fig. 7(a)–(e) shows the fracture surfaces of all the test materials after compressive loading. In all the test samples, fracture occurred at ∼45◦ angle with respect to the compression test axis. The fractographic evidences indicated that pure Mg exhibited dominant shear failure and showed more shear bands when compared to other Mgbased materials, which on the other hand exhibited rough fracture surfaces with mixed mode of shear and brittle features. 4. Discussion 4.1. Synthesis Synthesis of pure Mg and the Mg-based materials was successfully carried out by the DMD process followed by hot extrusion. The visual observation of as-cast and extruded samples indicates the absence of any macro-defects which clearly designates the suitability of processing parameters used in this study [10,24]. There was no reaction between the Mg melt slurries with the graphite crucible [10,13], which attributes to the inability of Mg to form stable carbides. Also the inert Ar gas atmosphere has prevented the oxidation of Mg melt. The characterization results clearly indicate the feasibility of the DMD process as a potential fabrication technique for Mg based materials [24]. 4.2. Coefficient of thermal expansion Based on the results of CTE (Table 1), with respect to pure Mg, the slight increase in the thermal expansion coefficients of Mg–3Al–5.6Ti and Mg–3Al is due to the higher CTE value of Mg17 Al12 (∼29.3 × 10−6 /K) [25] when compared to pure Mg (∼27.1 × 10−6 /K) [1]. On the other hand, the decrease in CTE observed in Mg–5.6Ti and Mg–(3Al + 5.6Ti)BM is attributed to the
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
155
Table 4 Results of room temperature tensile testing. S. No.
Material
0.2 Yield strength (MPa)
Ultimate tensile strength (MPa)
Failure strain (%)
1 2 3 4 5 6 7 8 9 10
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti Mg–(3Al + 5.6Ti)BM Mg–15.4SiC [11] (DMD) Mg–1.1Al2 O3 [18] (DMD) Mg–(TiB2 + TiC) [19] (remelting and dilution) Mg–10Mg2 Si [20] (powder metallurgy) Mg–22 vol%Saffil [21] (melt infiltration)
125 ± 9 161 ± 15 158 ± 6 167 ± 5 194 ± 2 155 ± 1 209 ± 1 95 ± 2 225 –
169 ± 11 250 ± 20 226 ± 6 236 ± 5 265 ± 2 207 ± 9 242 ± 3 298 ± 2 – 163
6.2 ± 0.7 7.5 ± 1.1 8.0 ± 1.5 4.8 ± 0.5 4.8 ± 0.6 1.4 ± 0.1 3.5 ± 0.3 2.4 ± 0.4 1.0 1.0
lower CTE values of Ti (∼9.1 × 10−6 /K) [1] and Al3 Ti intermetallic phase (∼13 × 10−6 /K), respectively [26].
4.3. Microstructural characterization 4.3.1. Effect of the addition of soluble element (Al) Microstructural characterization of Mg–3Al (Table 2) shows that the addition of mutually soluble element (Al) refines the grain size. It should be noted that Al is a major alloying element in Mg due to its high solid solubility [2,3,6]. Hence, the addition of 3Al to Mg forms the Mg–3Al alloy. It is well-known that during the solidification of Mg–Al alloys, primary-Mg solid solution with fine grains would be nucleated, along with the formation of Mg17 Al12 eutectic network distributed along the grain boundaries [27]. Further, the hot deformation of the cast ingot is believed to result in the breakdown of the Mg17 Al12 network structure into fine particles [27]. In the present case, it is evident from Fig. 1 (XRD) that the Mg–3Al alloy is composed of (Mg + Mg17 Al12 ) and from Figs. 2(a) and 3(a) it can be seen that the extruded Mg–3Al alloy consists of fine grains (<10 × 10−6 m) with fine individual Mg17 Al12 particles of size < 5 × 10−6 m near the triple grain edges. The formation of Mg17 Al12 intermetallic phase at grain edges also prevents grain growth during recrystallization [28]. In addition, as the extrusion was performed at ∼623 K (which is >0.5Tm of the Mg metal), it results in recrystallization, thereby leading to the formation of nearly equiaxed grains (Table 2) [28].
4.3.2. Effect of the addition of insoluble element (Ti) When Mg is incorporated with Ti (Mg–5.6Ti), structural analyses indicate that no intermetallic phases are formed (Figs. 1 and 3(b)). Based on the Mg–Ti binary phase diagram, the solubility of Ti in Mg is negligible and it does not form any intermetallic phase with Mg [29]. However, the good wettability of Ti by molten Mg facilitates good interfacial bonding [30] and hence the addition of insoluble element Ti to Mg can be regarded as a metallic reinforcement, i.e. Mg–5.6Ti composite which is composed of (Mg + Ti).
From the microstructure of Mg–5.6Ti (Fig. 3(b)), it is seen that the distribution of Ti is fairly uniform. However, the Ti particles are irregular in shape (sharp-edged and unevenly sized particles, Fig. 4(a)) and due to the tendency of irregularly shaped particles to agglomerate [31,32], clustering of particles can be dominantly observed. The grain size analyses indicate that grain refinement is observed when compared to pure Mg, but not when compared to Mg–3Al alloy (Table 2). This can be attributed to the larger size of the Ti-particles. It has been reported by various authors that small-sized second phases/particles effectively inhibit grain growth during recrystallization and significantly refine the grain size, when compared to second phase/reinforcement with larger sizes [17,28]. Hence, in Mg–5.6Ti, the presence of larger Ti particles with an average size range of ∼20–40 × 10−6 m results in coarser grain size when compared to that in Mg–3Al. 4.3.3. Effect of combined addition of Al and Ti metallic elements 4.3.3.1. Thermodynamics of Al–Ti system. Considering the Al–Ti phase diagram [29], the predominant equilibrium intermetallic phases are Ti3 Al, Ti2 Al5 , AlTi, Al2 Ti and Al3 Ti, respectively. Thermodynamic estimations show that the free energy of formation of Al3 Ti is estimated to be −33 kJ/mol, and is the lowest when compared to AlTi and Ti3 Al, as Ti2 Al5 and Al2 Ti are considered as transitional phases [33]. 4.3.3.2. Microstructure of ball milled powder. In the present work, although the composition of the powders used for ball-milling lies in the compositional range of AlTi intermetallic phase formation (65%Ti–35%Al), based on the free energy estimates it is expected that Al3 Ti phase is likely to form during the solid-state reaction. Such a behaviour was reported by Yang et al. [33] in their work on Mg–Al–Ti during reactive sintering. According to Yang et al., due to the absence of chemical reactions between Mg and Ti, and based on the free energy considerations, the Al3 Ti intermetallic phase is preferentially formed during solid-state reactive sintering [33]. In the present case, the XRD analysis confirm the formation of Al3 Ti (Fig. 1) although the peaks were not prominent, and contains the
Table 5 Results of room temperature compression testing. S. No.
Material
0.2 Compressive yield strength (MPa)
Ultimate compressive strength (MPa)
Failure strain (%)
1 2 3 4 5 6 7 8 9 11 12
Pure Mg Mg–3Al Mg–5.6Ti Mg–3Al–5.6Ti) Mg–(3Al + 5.6Ti)BM Mg [21] Mg–22 vol%Saffil [21] RZ5 [21] RZ5-22 vol%Saffil [21] AZ91 + 10vol%SiC + 3wt.%Si (squeeze casting) [22] WE54-13 vol%SiC (PM) [23]
74 ± 3 127 ± 2 85 ± 3 104 ± 1 139 ± 6 – – – – 160 230
273 ± 11 437 ± 5 360 ± 5 378 ± 13 431 ± 14 169 339 308 445 350 370
22.7 ± 4.9 21.7 ± 4.3 13.6 ± 1.2 12.6 ± 1.3 12.9 ± 1.6 11.9 4.6 16.7 5.2 9 14
156
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
Fig. 6. Fractographs of Mg-based materials after tensile loading showing: (a) dominant cleavage fracture in pure Mg, (b) ductile and cleavage fracture in Mg–3Al, (c) ductile features in Mg–5.6Ti, (d) particle debonding in Mg–5.6Ti, (e) mixed mode fracture and particle debonding in Mg–3Al–5.6Ti and (f) microvoid coalescence and crack formation at the vicinity of the particle/matrix interface in Mg–(3Al + 5.6Ti)BM .
peaks of Ti which is due to the high wt.% of Ti. Several rod-like particles (marked by arrows) observed in the powder after ball-milling (Fig. 3(d)) also indicates the formation of Al3 Ti. Further, microstructural investigation of the ball milled powder reveal the change in the morphology of elemental particulates of Al and Ti after the ball milling process (Fig. 4(c)), with the sharp corners of the as-received particulates eventually being rounded due to ball milling. In addition, the distribution of number of particulates vs. particulate size also indicates the changes induced as a result of ball milling, such as the reduction in the size of larger Ti particulates (Fig. 4(d)). 4.3.3.3. Microstructure of Mg–(3Al + 5.6Ti)BM . Upon the incorporation of (3Al + 5.6Ti)BM powder in Mg, the bulk Mg–(3Al + 5.6Ti)BM specimen reveals the presence of Al3 Ti, as observed from the phase identification (Fig. 1, XRD). This is due to the non-reactivity of Mg and Ti [29], and owing to the free energy of formation
of Al3 Ti [33]. It should also be noted that no Mg17 Al12 phase was detected. The formation of Al3 Ti is also supported by SEM micrographs wherein several blocky or rod-shaped particles of size ∼10–50 × 10−6 m are evident. From earlier reports, the Al3 Ti usually forms as blocky, rod-shaped or needle-shaped particles, with the size varying between submicrons to several tens of microns [34–37]. Hence, the Mg–(3Al + 5.6Ti)BM can be considered as an intermetallic reinforced Mg-composite, which is composed of Mg + (Ti + Al3 Ti). In comparison, the X-ray diffraction pattern of Mg–3Al–5.6Ti, wherein Al and Ti have been added directly during Mg melting, shows the diffraction peaks of Mg17 Al12 indicating the formation of the eutectic phase (Fig. 1). The estimated free energy of formation of Mg17 Al12 phase (liquid state) reported in literature is found to be −40 kJ/mol at 1023 K [37], which is more negative than that of Al3 Ti [33]. Hence the direct addition of Al and Ti to Mg during
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
157
Fig. 7. Fractographs of Mg-based materials under compression showing dominant shear failure.
liquid-state processing favours the formation of Mg17 Al12 rather than that of Al3 Ti. Therefore, the direct addition of Al and Ti into Mg can be regarded as Mg–3Al composite with Ti as metallic reinforcement, i.e. Mg + (Ti + Mg17 Al12 ). Based on these observations, it is evident that the pre-processing of metallic elements (solid-state reactive milling) significantly influences the microstructure of the developed Mg-based materials. Further, when compared to the rest of the Mg-based materials developed, the addition of ball-milled powder to Mg, (i.e. in Mg–(3Al + 5.6Ti)BM composite), has resulted in the reduction of average grain size of the composite, which is evident from the microstructure and the image analysis results (Table 2, Figs. 2(e) and 4(d)). As mentioned earlier, ball milling significantly reduces the average size and shape of the particles, and increases the number of small sized particles (Fig. 4(c)). Thus, during recrystallization, the increase in the number of sites for grain nucleation (large
number of small sized particles), effectively reduces the matrix grain size [18]. Further, it should be noted that the change in morphology of the (3Al + 5.6Ti)BM powders due to ball milling also leads to lower porosity levels in Mg–(3Al + 5.6Ti)BM composite when compared to Mg–5.6Ti and Mg–3Al–5.6Ti. The relatively higher porosity levels observed in Mg–5.6Ti and Mg–3Al–5.6Ti can be attributed to the formation of metal free zones at the sharp corners of Ti particulates, as a result of the inability of high viscosity particulate-molten alloy slurry to negotiate the sharp corners [38]; this has been avoided in Mg–(3Al + 5.6Ti)BM due to the round edged morphology obtained in the particles after ball milling. 4.4. Mechanical properties Literature study reveals that the addition of alloying elements and reinforcements to pure Mg has successfully improved the strength properties [5,9]. However the strengthening effect
158
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
depends on the properties of alloying elements/reinforcements such as (i) morphology and distribution, (ii) high hardness and strength properties, (iii) ability for grain nucleation, (iv) obstruction to dislocation movements, and (v) effective load transfer from the matrix [39,40]. 4.4.1. Microhardness The microhardness measurements shown in Table 3 indicate an increase in the mean hardness values of the Mg based materials when compared to pure Mg. This increase in hardness values corresponds to the presence of comparatively harder second phases such as Ti in Mg–5.6Ti, Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM , Mg17 Al12 in Mg–3Al and Mg–3Al–5.6Ti and Al3 Ti in Mg–(3Al + 5.6Ti)BM . Although Mg–3Al has finer grain size and that the Mg17 Al12 phase has higher hardness than Ti (Mg17 Al12 ∼2.2 GPa and Ti ∼0.97 GPa [41,42]), the hardness value of Mg–3Al alloy is slightly lower than that of Mg–5.6Ti. This can be attributed to the relatively higher volume fraction of the sharp-edged Ti particulates and their distribution, as sharp-edged particulates tend to increase the stress concentration at particle corners resulting in increased dislocation density, thereby contributing to higher hardness values [43]. Among all the materials, the maximum mean hardness value observed is in the Mg–(3Al + 5.6Ti)BM composite, which is due to the inherent high hardness of Al3 Ti (∼4 GPa) [44]. 4.4.2. Tensile and compressive properties In the current investigation, the results from room temperature tensile and compression tests indicate a significant improvement in the strength properties of Mg incorporated with metallic elements. The effect of these additions on the mechanical properties has been discussed in the following sections. 4.4.2.1. Mg–3Al alloy. In the current work, the improvements in both strength and ductility obtained due to the addition of Al (Mg–3Al alloy) in tensile loading and improvement in strength and minimal adverse effect on ductility under compressive loading condition can be seen in Tables 4 and 5 and Fig. 5. As mentioned earlier, Al is one of the major alloying elements known to improve the strength properties of Mg, and due to its high solubility in Mg, it forms the eutectic Mg17 Al12 second phase [3,27]. Based on the microstructural analysis, it is evident that the addition of Al refines the grain size significantly. Further, due to the hot extrusion process, the presence of small sized individual Mg17 Al12 particles occur along the grain boundary, mostly at triple grain edges, rather than as eutectic network. Hence, in the Mg–3Al alloy the improvement in strength properties can be attributed to the following strengthening mechanisms: (i) strengthening due to the grain size reduction, due to the formation of refined and equiaxed grains [45], (ii) solution strengthening effect [28] and (iii) formation of Mg17 Al12 phase at the grain edges that impede dislocation motion [27]. Further, the lower Al content, the reduction in grain size and the presence of small sized second phase particulates have resulted in the improvement in ductility. Tensile fracture surface morphology of Mg–3Al alloy confirms the dominant ductile fracture due to plastic deformation (Fig. 6(b)) when compared to the fracture surface of pure Mg, which shows fracture by cleavage (Fig. 6(a)). Compressive fracture surfaces of both pure Mg and Mg–3Al alloy shows failure by shear band formation (Fig. 7(a) and (b)). 4.4.2.2. Mg–5.6Ti composite. From Tables 4 and 5, the improvement in mechanical properties of Mg when incorporated with Ti particulates (Mg–5.6Ti composite) is evident under both tension and compression, though the values are lower than Mg–3Al. Also, the Mg–5.6Ti composite exhibits strength and ductility values
that are similar to or higher than those of ceramic reinforced Mgcomposites, a few of which are listed in Tables 4 and 5. These results indicate that the addition of Ti particles strengthen the composite by efficient load transfer. Based on the microstructural observations and the inherent properties of Ti particulates, the improvement in mechanical properties when compared to pure Mg can be ascribed to the following strengthening effects: (i) the increase in the dislocation density due to the thermal residual stresses, which arises from the mismatch in the coefficient of thermal expansion between Mg and Ti, contribute to the increase in the yield strength, (ii) strength improvement due to grain refinement, (iii) the morphology of the Ti-particles, wherein the sharp-edged particles contribute to higher dislocation density due to the increased stress concentration at the pointed corners [43], and (iv) higher hardness of Ti which eventually increases the load carrying capacity (ultimate strength) [39,40,45]. While the higher inherent tensile fracture strain of Ti [46] increases the tensile fracture strain, due to its low strain value under compression [47], the fracture strain of Mg–5.6Ti has reduced. In comparison, Mg with other metallic reinforcements like Ni or Cu reported earlier [12,13] show low tensile ductility, which is due to the chemical activity between Mg and Ni/Cu forming brittle intermetallics with Mg such as Mg2 Cu and Mg2 Ni. In the present case, the improved behaviour (both strength and ductility) of Mg–5.6Ti is not only due to the absence of formation of Mg-based intermetallic phases [29], but also due to the good wettability and structural compatibility of Ti and Mg (HCP structure) [30]. Tensile fracture surface of Mg–5.6Ti show mixed mode fracture (ductile + cleavage) modes (Fig. 6(c) and (d)). As the hard Ti particles efficiently aid in the load bearing capacity, they resist crack propagation and thereby increase the tensile ductility [14]. However, as Ti is insoluble in Mg (chemical inertness), the matrix–particle interface would be predominantly governed by mechanical bonding due to the good wettability between Ti and Mg, rather than chemical bonding [48]. Hence when the maximum load is reached, rather than particle fracture, debonding of Ti particulates would occur (Fig. 6(d)). Under compressive loads, while shear failure is the dominant mode of fracture (shear + brittle mode), the fracture surface appears relatively rough when compared to pure Mg (Fig. 7(a) and (c)). 4.4.2.3. Mg–(3Al + 5.6Ti)BM composite. Based on the tensile and compression test results (Tables 4 and 5), it has been observed that when a combination of soluble Al and insoluble Ti is incorporated into pure Mg after pre-processing, the resulting Mg–(3Al + 5.6Ti)BM composite shows enhanced strength properties, at the expense of ductility. When compared to pure Mg, ∼50% increase in yield and tensile strengths are evident, and under compression, the yield strength has been doubled (∼100% increase) with significant increase in ultimate strength as well. The strengthening in Mg–(3Al + 5.6Ti)BM composite occur due to the following reasons: (i) presence of hard Ti particulates [41], (ii) formation of hard Al3 Ti intermetallic phase [44], (iii) significant grain refinement and (iv) the increase in the dislocation density due to the thermal residual stresses, owing to the mismatch in the CTE between Mg and Al3 Ti [1,26]. These together contribute towards the increase in the strength properties [39,40]. Although intermetallic phases such as Al3 Ti tend to reduce the ductility, they are considered to be advantageous over conventional ceramic reinforcements [49]. It has been reported that the CTE values of intermetallics are closer to that of metals and are higher than that of ceramics. Thus they reduce the residual stresses at the particle/matrix interface relatively, thereby improving the ductility (i.e. ductility loss is relatively less when compared to ceramic reinforcements) [49]. This is evident from Tables 4 and 5, wherein a comparison made with Mg-composites
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
containing ceramic reinforcements show drastic reduction in ductility. Further, the ductility values of both Mg–3Al–5.6Ti and Mg–(3Al + 5.6Ti)BM composites are similar, in spite of the presence of Al3 Ti intermetallic phase in Mg–(3Al + 5.6Ti)BM . This is primarily due to the change in the morphology of the particulates. It has been reported that for composites containing angular or sharpedged particles, the stress concentration would occur at the pointed corner of the particle, and that the presence of such particles in the composite reduces the ductility of the composite [43]. Eliminating such particle corners is reported to render uniform stress distribution and improves the ductility without decreasing the strengthening effect [43]. Hence, in spite of the presence of brittle Al3 Ti intermetallic phase in Mg–(3Al + 5.6Ti)BM , the presence of particulates with rounded edges in Mg–(3Al + 5.6Ti)BM composite would lead to reduced stress concentration and uniform stress distribution, thereby exhibiting better ductility. The tensile fracture surface morphology of Mg–(3Al + 5.6Ti)BM composite showed mixed mode fracture features. From Fig. 6(f), it can be seen that the interface bonding is good but there are also regions at the particle/matrix interface that contains microvoids that coalesce to form cracks that extend into the matrix. Chawla and Chawla [50] reported that in discontinuously reinforced metal matrix composites with strong interface, the matrix is significantly work hardened due to the presence of the hard reinforcements. As a result, the matrix would be placed under a very large degree of constraint with an inability for strain relaxation to take place by deformation. This eventually results in the onset of void nucleation and propagation at the particle/matrix interface, as is evident from the microvoids and cracks seen in the present case. In contrast, in the Mg–3Al–5.6Ti composite, particle debonding was dominant (Fig. 6(e)). Under compression, both composites show failure by shear fracture, but the surfaces appear very rough when compared to other Mg-materials (Fig. 7(d) and (e)). 5. Conclusions The following are the conclusions that could be drawn from the present study: (1) DMD technique can successfully synthesize magnesium alloys and composites. (2) The addition of soluble element (Al) in lower content formed the Mg–3Al alloy with Mg17 Al12 eutectic phase that refined the grain size and improved the hardness, tensile and compressive strength properties. (3) The addition of insoluble element (Ti) formed the Mg–5.6Ti composite. Based on the inherent properties of Ti such as hardness, ductility, structural compatibility with Mg and the good wettability of Ti and Mg, the Mg–5.6Ti composites exhibited enhancement in strength and ductility, especially under tensile loads. (4) The effects of the addition of the combination of soluble and insoluble metallic elements primarily depend on their method of addition. (5) While the pre-processing of Al and Ti by ball-milling initiated solid state reaction resulting in Ti + Al3 Ti intermetallic phase, the direct addition of elements to Mg resulted in Ti + Mg17 Al12 phase formation. Hence the Mg–(3Al + 5.6Ti)BM composite was composed of Mg + (Ti + Al3 Ti) while that of Mg–3Al–5.6Ti was made of Mg + (Ti + Mg17 Al12 ). (6) The Mg–(3Al + 5.6Ti)BM composite showed significant grain refinement, lower CTE and substantial improvement in hardness, tensile and compressive strengths, with a loss in ductility. These can be attributed to the inherent properties of Al3 Ti intermetallic phase and a change in morphology of the
159
particles. In comparison, the Mg–3Al–5.6Ti composite showed lower strength properties. (7) With pure Mg as the matrix, a combination of superior mechanical properties can be achieved by careful selection and incorporation of metallic elements and their method of addition. Acknowledgements This work has been supported in part by the Agency for Science, Technology and Research (ASTAR) grant 092 137 0015 (WBS# R265-000-321-305). One of the authors, Mr. S. Sankaranarayanan acknowledges the NUS research scholarship for supporting this research for his graduate program. References [1] C.J. Smithells, Metals Reference Book, 5th ed., Butterworth’s & Co. Ltd., London, 1976. [2] E.F. Emley, Principles of Magnesium Technology, Pergamon Press, Oxford, 1966. [3] M.M. Avedesian, H. Baker, ASM Specialty Handbook. Magnesium and Magnesium Alloys, ASM International, OH, 1999. [4] G. Pettersen, Mater. Sci. Eng. A 207 (1996) 115–120. [5] W. Du, Y. Sun, X. Min, F. Xue, M. Zhu, D. Wu, Mater. Sci. Eng. A 356 (2003) 1–7. [6] I.J. Polmear, Mater. Sci. Technol. 10 (1994) 1–16. [7] A. Luo, M.O. Perquleryuz, J. Mater. Sci. 29 (1994) 5259–5271. [8] C. Fritze, H. Berek, K.U. Kainer, S. Mielke, B. Wielage, in: B.L. Mordike, K.U. Kainer (Eds.), Magnesium Alloys and Their Applications, Werkstoff-Informationsgesellschaft, Frankfurt, Germany, 1998, pp. 635–640. [9] K.U. Kainer, Metal Matrix Composites Custom Made Materials for Automotive and Aerospace Engineering, Wiley, 2006. [10] M. Gupta, M.O. Lai, D. Sarvanaranganathan, J. Mater. Sci. 35 (2000) 2155–2165. [11] S. Ugandhar, M. Gupta, S.K. Sinha, Compos. Struct. 72 (2006) 266–272. [12] S.F. Hassan, M. Gupta, J. Alloys Compd. 335 (2002) L10–L15. [13] S.F. Hassan, M. Gupta, Mater. Res. Bull. 37 (2002) 377–389. [14] P. Peˇırez, G. Garceˇı s, P. Adeva, Compos. Sci. Technol. 64 (2004) 145–151. [15] S.F. Hassan, M. Gupta, J. Alloys Compd. 345 (2002) 246–251. [16] C. Suryanarayana, Mechanical Alloying and Milling, CRC Press, New York, 2004. [17] M.K. Habibi, S.P. Joshi, M. Gupta, Acta Mater. 58 (2010) 6104–6114. [18] S.F. Hassan, M. Gupta, J. Alloys Compd. 419 (2006) 84–90. [19] X. Zhang, H. Wang, L. Liao, X. Teng, N. Ma, Mater. Lett. 59 (2005) 2105–2109. [20] L. Lu, K.K. Thong, M. Gupta, Compos. Sci. Technol. 63 (2003) 627–632. [21] D.J. Towle, C.M. Friend, Mater. Sci. Technol. 9 (1993) 35–41. [22] Z. Trojanova, V. Gartnerova, A. Jager, A. Namesny, M. Chalupova, P. Palcek, P. Lukac, Compos. Sci. Technol. 69 (2009) 2256–2264. [23] Z. Szaraz, Z. Trojanova, M. Cabbibo, E. Evangelista, Mater. Sci. Eng. A 462 (2007) 225–229. [24] S. Sankaranarayanan, S. Jayalakshmi, M. Gupta, J. Alloys Compd. 509 (2011) 7229–7237. [25] V. Goryany, P.J. Mauk, J. Miner. Metall. 43 (2007) 85–97. [26] T. Li, E.A. Olevsky, M.A. Meyers, Mater. Sci. Eng. A 473 (2008) 49–57. [27] A. Kielbus, T. Rzychon, R. Chibis, J. Achiev. Mater. Manuf. Eng. 18 (2006) 135–138. [28] Z. Li., J. Dong., X.Q. Zeng., C. Lu., W.J. Ding., Mater. Sci. Eng. A 466 (2007) 134–139. [29] Binary Alloy Phase Diagram [Electronic Resource], 2nd ed., Plus Updates, ASM International, Materials Park, OH, 1996. [30] K. Kondoh, M. Kawakami, H. Imai, J. Umeda, H. Fujii, Acta Mater. 58 (2010) 606–614. [31] Z. Wang, M. Song, C. Sun, Y. He, Mater. Sci. Eng. A 528 (2011) 1131–1137. [32] A. Slipenyuk, V. Kuprin, Yu. Milmana, J.E. Spowart, D.B. Miracle, Mater. Sci. Eng. A 381 (2004) 165–170. [33] Z.R. Yang, S.Q. Wang, M.J. Gao, Y.T. Zhao, K.M. Chen, X.H. Cui, Composites A 39 (2008) 1427–1432. [34] X. Wang, A. Jha, R. Brydson, Mater. Sci. Eng. A 364 (2004) 339–345. [35] Y. Watanabe, Y. Fukui, Multidiscip. Microsc. Res. Edu. 18 (2004) 9–198. [36] Y.-h. Zhao, J. Zhou, W. Yan, Trans. Non-ferrous Met. Soc. China 12 (2002) 643–648. [37] D.W. Zhou, J.S. Liu, S.H. Xu, P. Peng, Physica B 405 (2010) 2863–2868. [38] M. Gupta, M.O. Lai, C.Y. Soo, Mater. Res. Bull. 30 (1995) 1525–1534. [39] D.J. Lloyd, Int. Mater. Rev. 39 (1994) 1–23. [40] I.A. Ibrahim, F.A. Mohamed, E.J. Lavernia, J. Mater. Sci. 26 (1991) 1137–1156. [41] ASM Handbook Properties and Selection: Non-Ferrous Alloys and SpecialPurpose Materials, vol. 2, ASM International, Materials, Park, OH, 1990. [42] M.S. Yoo, J.J. Kim, K.S. Shin, N.J. Kim, in: H.I. Kaplan (Ed.), Magnesium Technology, The Minerals, Metals & Materials Society, TMS, 2002. [43] S. Qin, C. Chen, G. Zhang, W. Wang, Z. Wang, Mater. Sci. Eng. A 272 (1999) 363–370. [44] Yu.V. Milmana, D.B. Miracleb, S.I. Chugunovaa, I.V. Voskoboinika, N.P. Korzhovaa, T.N. Legkayaa, Yu.N. Podrezova, Intermetallics 9 (2001) 839–845. [45] G.E. Dieter, Mechanical Metallurgy, McGraw-Hill, USA, 1986.
160
S. Sankaranarayanan et al. / Materials Science and Engineering A 530 (2011) 149–160
[46] Metals Hand Book – Properties and Selection of Metals, 8th ed., American Society for Metals, Metals Park, OH, p. 1225. [47] P.F. Barrett, Compressive Properties of Titanium Sheet at Elevated Temperatures, Technical Note 2038, National Advisory Committee for Aeronautics.
[48] S. Ahmed, F.R. Jones, Composites 21 (1990) 81–84. [49] Z.R. Yang, S.Q. Wang, X.H. Cui, Yu.T. Zhao, M.J. Gao, M.X. Wei, Sci. Technol. Adv. Mater. 9 (2008). [50] N. Chawla, K.K. Chawla, Metal Matrix Composites, Springer, New York, 2006.