Processing and mechanical properties of pure mg and in situ aln reinforced mg 5al composite 4

Processing, physical and tensile properties Chapter Processing, physical and tensile properties 4.1 Introduction Due to ease of fabrication at relatively low cost, conventional metal-matrix composites reinforced with ceramic particulates are becoming the preferred choices of structural materials. They exhibit high strength and elastic modulus, near-isotropic as well as excellent high-temperature creep resistant properties. The failure mode, strength, and ductility of composites vary with particulate size. Increase in particle size results in a decrease in both tensile strength and ductility. During mechanical loading, large ceramic particulates have high tendency to cracking that leads to premature failure and low ductility of the composites. This can be avoided by using smaller ceramic particulate size. Further enhancement in mechanical properties can be realized by developing nanocomposites in which reinforcement particles and/or matrix grains are in nanometer regime [1-7]. Severe plastic deformation and MA/MM processes can be employed to refine the matrix grains [8,9]. MM is one of the most effective processes for dispersing ex-situ nanoparticles more uniformly in metal matrix [5,6,10-14] and inducing in-situ nanoparticles in the composites during milling. A better bonding between metal matrix and in-situ formed nanoparticles which are clean, ultrafine and thermally stable renders the excellent mechanical properties. Inherent deficiencies such as low stiffness, high wear rate, and high chemical reactivity, loss of mechanical strength at high temperature and creep resistance restrict the 69 Processing, physical and tensile properties industrial applications of Mg and its alloys [15]. By adding micro and nanosized ceramic particles in Mg matrix, these drawbacks can be overcome [3, 6]. In this study, Mg nanocomposites with wt% of in-situ AlN formed reinforcement were synthesized for milling durations up to 40h and their physical and mechanical properties were accessed. The contribution of texture developed during extrusion to tensile deformation was examined by means of pole figure measurements. For comparative study, pure Mg samples were also synthesized and tested using identical parameters used for composite samples. 4.2 Experimental Mg chips/turnings (Drehspaene) (Acros Organics) and Al powder (Alfa Aesar, -325 mesh) of 99.5% and 99% purity respectively were used as starting materials. AlN composite powder was synthesized in-situ by MM of Al powder and pyrazine for 100h as described in Chapter 3. The nominal composition of the composite is Mg-5wt%Al1wt%AlN (Mg-5Al-1AlN). 35g of composite mixture together with 0.5 to wt% of stearic acid, CH3(CH2)16COOH, and hardened carbon steel balls were loaded into 500 ml stainless steel vial in a 99.9% pure argon atmosphere in an AMBRUAN glove box. The weight ratio of Mg chip to ball is 1:20. A Retsch PM400 Planetary Ball Mill was employed for MM at 300 rpm. Each batch of powder was mechanically milled for different durations of (as-blended), 10, 20, 30 and 40 hours at room temperature. 0hMMed sample was obtained by blending the composite mixture at low rotational speed of 100 rpm for 1h. Same milling conditions were applied to MM of pure Mg chips. Mg-5Al-1AlN and pure Mg samples are designated hereafter as xxh-MMed composite sample and xxh-MMed Mg sample respectively, where xx is milling hours while the as-blended powder mixture or as-received Mg chips are indicated as 0h. 70 Processing, physical and tensile properties After milling, a small quantity of powder was withdrawn for the examination of structural changes by means of an X-ray diffractometer (XRD). The milled powders were cold-compacted using 35mm diameter metal die at sixty tons of uniaxial compaction pressure. The green compacts were sintered in a tubular furnace under argon gas flow for hours at 500ºC. The sintered billets were then hot-extruded at an extrusion ratio of 25:1 to cylindrical rods of 7mm diameter. The grain size of the as-received Mg chips and as-blended extruded specimens was measured using optical microscope and the microstructure of as-milled specimens were characterized using Jeol 2010F TEM. The extruded rods were machined into cylindrical tensile specimens with a gauge diameter of 5mm and a gauge length of 25 mm according to ASTM E8M-96 standard. Uniaxial tensile test was conducted at room temperature using an automated Instron 8501 servo hydraulic testing machine at controlled strain rates of 3.33x10-4 s-1. The deformation was monitored using a 25-mm clip-on extensometer. Resistivity measurement was carried out by Jandel Multi Height Four-Point Probe Stand with Keithley K6200 DC current source and Keithley K2182 nanovoltmeter. The bulk resistivity  was obtained from the equation:    2s  V I (4.1) where s is the spacing of the probe in cm, I the test current in ampere and V the measured voltage in volt. 71 Processing, physical and tensile properties Setaram TMA 92-16.18 was employed to investigate the nature of thermal expansion of the samples by means of coefficient of thermal expansion (CTE). Thermal behavior of the bulk sample was further investigated by heating the samples in differential scanning calorimeter DSC-2910 to 700°C at 10°C/min. To calculate the specific heat capacity, thermal analysis was carried out using DSC from 323K to 453K at a constant heating rate of 20 K/min in argon atmosphere. Specific heat capacity Cp,sample was obtained from equation 4.2 [16]. C p , sample  msapphire ysample  reference msample ysapphire  reference C p , sapphire (4.2) where Cp,sample and Cp,sapphire, msample and msapphire are the specific heat capacities and weights of measured sample and a sapphire standard sample. The effective displacement of the sample Δysample-reference and the sapphire standard Δysapphire-reference are the difference between the distances from the reference baseline in the thermal plot. The standard specific heat capacity of sapphire was obtained from the thermodynamic data [17] and is expressed as C p ,sapphire  104.547  2(12.087 x10 3 T )  2(1.757 x10 T 2 )  6(0.742 x10 6 T ) ( 474.635 0.5 T ) (4.3) 4.3 Results and discussion 4.3.1 Mass structure investigation by XRD X-ray diffractometer was employed to perform structural investigation on MMed powders and extruded specimens in the transverse direction. XRD spectra of the MMed composite powders and the extruded composite samples are shown in Figs. 4.1(a) and (b) respectively. In Fig. 4.1(a), all Al peaks from the as-blended specimen 72 Processing, physical and tensile properties disappeared in the MMed powder due to solid solution of Al with Mg resulting in the formation of Al12Mg17. A new phase of MgAl2O4 was detected in the MMed powder after 10 and 20h. During milling, MgO and Al2O3 oxide layers on as-received Mg chips and Al power surfaces fractured into very fine particles to favor solid state reaction for the formation of MgAl2O4 according to the following reaction [18]. Mg + Al2O3 3 MgAl2O4 + Al (4.4) Very weak AlN peaks were observed in all MMed specimens suggesting the complete immiscibility of AlN in Mg. With increasing milling duration, broadening of XRD peaks and declining in peak intensity are observed due to the reduction in grain size and introduction of microstrain during milling. In Fig. 4.1(b), intensities of Mg (100) and (110) peaks for the extruded specimens increase with milling duration, confirming the formation of deformation texture. Texture occurs due to the deformation-induced crystallographic plane rotating preferentially along the extrusion direction as a result of extrusion at high extrusion ratio. Contamination from process control agent, stearic acid and powder handling atmosphere can also be observed from the weak MgO (200) peak and MgH2 (211) peak, especially at milling durations of 20h and longer. Al12Mg17 peaks disappeared in all the as-milled specimens. It might be due to very fine in size and very minimal in quantity to produce a visible diffraction peak. It is also highly possible that Al reacted with the excess nitrogen molecules from AlN composite powder to form AlN. This will be confirmed by thermal analysis of the composite samples by DSC in section 4.3.3. Weak MgAl2O4 peaks emerged in all as-milled samples as a result of solid state reaction during sintering between MgO and Al2O3, both of which are inherited from the surfaces of the as-received Mg chips and Al powders. 73 20 h Mg Al12Mg17 AlN MgAl2O4 MgO (103) (112) (110) (102) (002) 10 h 0h 40 50 60 2 (d egree) (a) 70 80 40 h 20 h 10 h 30 MgH2 30 h Intensity (a.u.) 40 h 30 h Intensity (a.u.) (100) Al Al12Mg17 MgAl2O4 (112) (103) (110) (102) (100) (002) (101) Mg AlN (101) Processing, physical and tensile properties 0h 30 40 50 60 70 80 2 (d egree) (b) Figure 4.1 X-ray diffraction patterns of Mg-5Al-1AlN composite samples at different milling durations for (a) MMed powers and (b) MMed+extruded samples. From X-ray diffraction patterns of pure Mg in Figs. 4.2 (a) and (b), no apparent contamination from stearic acid or milling and handling atmosphere was observed in as-milled powders and as-extruded samples. It might be due to negligible amount of contamination or the particles of contamination by-products were too fine to produce prominent diffraction peaks. The intensity of Mg (100) peak was exceptionally high in the as-extruded specimens and that of basal plane (002) peak becomes lower with milling duration. It implies that extrusion induced deformation textures exist in the extruded samples. 74 Mg (112) (103) (110) (102) (112) (110) (102) (103) (100) (002) (101) Mg (100) (101) (002) Processing, physical and tensile properties 40 h 40 h 30 h Intensity (a.u.) Intensity (a.u.) 30 h 20 h 20 h 10 h 10 h 0h 0h 30 40 50 60 2 (d egree) (a) 70 80 30 40 50 60 70 80 2 (d egree) (b) Figure 4.2 X-ray diffraction patterns of pure Mg samples at different milling durations for (a) MMed powers and (b) MMed+extruded samples. Fig. 4.3(a) shows the microstructure of the as-received Mg chips. In Fig. 4.3(b), Al12Mg17 decorated along the Mg matrix grain boundaries and AlN along the Mg chip boundaries in the 0h-MMed extruded composite sample. Deformation due to cold compaction and hot extrusion was not high enough to inject AlN particles into the grain boundaries. From Fig. 4.3(c), no apparent grain elongation could be observed along the extrusion direction. Fig. 4.3(d) shows the clean grain boundaries of pure Mg samples. 75 Processing, physical and tensile properties 15 m 100 m (b) (a) 30 m 15 m (c) (d) Figure 4.3 Optical micrograph of (a) as-received Mg chip, (b) 0h-MMed composite sample in cross-sectional area, (c) 0h-MMed composite sample in longitudinal direction and (d) 0h-MMed Mg sample in longitudinal direction. At higher magnification, entangled dislocation pile-ups within the grain were observed from TEM image as shown in Fig. 4.4 (a). During milling, mechanically cold-worked powders resulted in generation of dislocations, multiplication and congealing that produced nanosized grains [19]. The grains were highly strained and contained numerous defects. When the grain is extremely small, the formation of new nanocrystals via dislocation movement stops because of inability of individual grain to support more than one dislocation [20]. As such, some limited dislocations (Figs. 4.4 (b) and (c)) can be observed in the 10h- and 20h-MMed composite samples which are in larger grain size regime. However, for 30h- and 40h-MMed composite samples, no 76 Processing, physical and tensile properties (a) (b) (c) (d) d100=2.75Å (e) (f) Figure 4.4 TEM images of (a) 0h-, (b) 10h-, (c) 20h-, (d) 30h-, (e) 40h-MMed composite sample and (e) AlN reinforcement (marked with dotted line) in 30h-MMed composite sample. 77 Processing, physical and tensile properties indication of the presence of dislocation can be observed in Figs. 4.4 (d) and (e). Fig.4.4 (f) shows a HRTEM image of lattice pattern from 30h-MMed composite sample. It reveals that the particle is AlN polycrystal with an inter-planer spacing of 2.75 Å corresponding to the (100) plane of AlN crystal. 40h-MMed composite showed the smallest grain size of 33 nm compared to the coarse and ultrafine grain-sized samples. In Fig. 4.5, HRTEM investigation reveals the estimated thickness of grain boundary was about nm and it appears to be free of contamination or particles with disordered phases. Grain boundary Figure 4.5 HRTEM observation of grain boundary marked with dotted line. As shown in Table 4.1, after 10h of milling, grain size of the MMed powders was significantly reduced from 24 m to 44 and 41 nm in the composite sample and pure Mg sample respectively. However, longer milling did not produce further grain refinement. The crystalline sizes are respectively 32, 26 and 22 nm after 20, 30 and 40h-MM in composite sample. In pure Mg samples, average crystalline sizes after 20, 30 and 40h-MM are 31, 28 and 25 nm respectively. 78 Processing, physical and tensile properties reinforcement. Based on Taylor dislocation strengthening mechanism, σEM and σCTE can be determined as:  EM  3  mb GEM (4.13)  CTE  3 mb GCTE (4.14) and where α and β are the strengthening coefficients, μm is the shear modulus of the matrix and b the Burgers vector. 4.3.5 Effects of texture on mechanical properties The main deformation mode in magnesium and magnesium alloys is basal slip, i.e. slip on the (0001) plane with a 112 0 Burgers vector. Prismatic slip {10 0} 112 0 and pyramidal slip {10 1} 11 0 have also been observed, but their critical resolved shear stress at room temperature is very much higher than that for basal slip [37]. The deformation texture in hcp metals and their alloys will develop in accordance with the relative contributions from the above deformation paths as well as twinning slip {10 2} 10 0 . Mukai et al. [38] has reported that two AZ31 extruded samples with almost the same grain size (~15µm) showed different tensile properties owing to the difference in texture. The sample with larger fraction of basal planes along the extrusion direction (tensile axis) produced higher YS and lower ductility. The (0002) pole figures of composite and pure Mg extruded samples with their reflecting surface normal to the extrusion direction shown in Figs. 4.13 and 4.14 displayed the typical basal fiber 92 Processing, physical and tensile properties (a) (b) (c) (d) (e) Figure 4.13 (0002) texture of as-extruded (a) 0h-, (b) 10h-, (c) 20h-, (d) 30h- and (e) 40h-MMed Mg-5Al-1AlN composite samples. 93 Processing, physical and tensile properties (a) (b) (c) (d) (e) Figure 4.14 (0002) texture of as-extruded (a) 0h-, (b) 10h-, (c) 20h-, (d) 30h- and (e) 40h-MMed pure Mg samples. 94 Processing, physical and tensile properties texture which tends to form in hcp structure with ideal c/a ratio of 1.633 (c/a ratio of Mg: 1.624) [39]. In this study, 30h-MMed composite sample and 20h-MMed pure Mg sample with highest intensity basal texture as shown in Figs. 4.13 (d) and 4.14 (c) should produce highest YS with lowest ductility. However, this theoretical prediction does not agree with the experimental results. Highest YS for the 20h-MMed composite sample and 10h-MMed pure Mg sample were observed while highest ductility was obtained after 40h-MM in both material systems. When the grain size decreases especially to nanoscale, increasing contribution from multiple slip, strong decrease in work hardening and GBS diminish the texture strengthening [40]. This could be related to the changes in the activity of the slip systems and deformation mechanism. In nanocrystalline materials, there exists a critical minimum grain size where any value below it, the grain can no longer support the dislocation and the traditional Hall-Petch relation based on the concept of dislocation pileups could no longer be applied. When the grain size becomes very small, the diffusion lengths required for grain orientation would become very short and grain boundary process such as diffusion may prevail. Also, in the nano-regime, dislocation activity such as gliding may decrease or even disappear and creation of new dislocations is very difficult [41]. Absence of strengthening contributed by grain refinement and reinforcement particles suggests a change in deformation mechanism from dislocation controlled plastic flow to grain boundary activity-controlled mechanism. 95 Processing, physical and tensile properties 4.3.6 Deformation mechanism Bulk nanostructured Mg composite as well as pure Mg samples exhibit several remarkable mechanical behaviors. It includes dramatic increase in strength after 10h and 20h of milling and decreased in strength but enhanced ductility with extensive elongation after 40h of milling. Subjected to micro-forging, fracture, agglomeration, and deagglomeration during MM process and deformation during hot extrusion, grain size of composite samples reduced dramatically from 17µm to 116 nm and 86 nm after 10h and 20h MM respectively. In addition to reinforcement AlN and Al solid solution in Mg, small volume fraction of second particles from contamination resulted from milling medium and atmosphere such as MgH2, MgAl2O4 and MgO were found in the milled samples. Based on the fact that pure Mg exhibited significant (155%) increase in YS compared to as-blended sample, the increase in strength of the 10h-MMed sample is believed to be predominantly due to grain refinement and lesser extent due to particle strengthening. Further milling to 30h and 40h produced lower YS but with enhanced ductility. After longer milling duration, Mg grain becomes much smaller (in nanometer range) and it is difficult for Al solute to infuse through Mg grain thus decreasing in solid solution of Al in Mg. Therefore, Al solute could have become part of the nano-grain boundary as suggested by Rawers et al. [19] leading to the weakening of solid solution strengthening rather than mechanically infused throughout the Mg grain to produce solid solution strengthening. It is observed that grain refinement due to reduction in grain size with further milling failed to increase the strength. 96 Processing, physical and tensile properties From TEM microstructure shown in Fig. 4.15 (a), dislocation pileups and entanglement were found in the unmilled composite sample after tensile deformation. However, no dislocation activities and apparent grain elongation could be detected in the deformed 30h- and 40h-MMed composite samples as shown in Figs. 4.15 (b) and (c). (a) (b) (c) Figure 4.15 (a) Bright field TEM of deformed as-blended composite sample in which dislocation pile-ups and entangled dislocations were clearly seen. Deformed (b) 30hand (c) 40h-MMed composite sample with poorly defined grain boundaries, no obvious grain elongation and dislocation activities. 97 Processing, physical and tensile properties The best-known effect of mean grain size on low-temperature mechanical properties is described by the Hall-Patch empirical relationship: [42,43]   M ( i  k HP d 1 / ) (4.15) where  is yield stress, M the Taylor factor, i the lattice frictional stress needed to move the dislocations, kHP the microstructure stress intensity and d the grain size. In this equation, the yield stress is approximately equal to the frictional stress i when the grain size d approaches infinity. Only when the grain size is sufficiently small, the grain boundary strengthening effect will be significant. In this Hall-Petch relationship, the strengthening mechanism is based on dislocation pile-ups at physical obstacles such as grain-boundaries. One of the rationalizations of equation 4.15 is that a pile-up at a grain boundary in one grain can generate sufficiently large stresses to operate sources in an adjacent grain at the yield stress. Dislocation pile-ups produce large, long-range stress at grain boundaries; this can nucleate either yielding in the adjacent grains or boundary cracks [44]. The critical stress c experienced by the leading dislocation for source operation is given by  c  n(   i )  n e   e2 d Gb (4.16) where  is the applied shear stress and i the lattice frictional stress, e the effective shear stress or applied resolved shear stress, n the number of dislocations, G the shear modulus and b, the Burgers vector. Equation 4.16 shows that the critical shear stress depends on the number of dislocations at the tip of pile-up and the grain size. When grain size becomes smaller the number of dislocation at the tip of pile-up is also 98 Processing, physical and tensile properties reduced and the applied resolved shear stress has to be increased so that dislocation source in the adjacent grain will be activated to cause dislocation slip. After the initial milling of 10h, grain size became smaller compared to unmilled sample and the number of dislocation pile-ups was limited and thus resulting in an increase in YS. Further reduction may lead to the spacing between dislocations comparable to the grain size d eventually and individual grain will no longer be able to support more than one dislocation [45]. In such situation, the contribution from grain boundary strengthening will be somehow different and there will be critical yield stress at certain grain size. Nieh and Wadsworth [45] have estimated the critical grain size by equating the repulsive force between two edge dislocations and the applied shear stress as follows: lc  Gb MGb   (1   )  (1   ) (4.17) where  is the Poisson's ratio and =M, the uniaxial tensile stress which is related to the critical resolved shear stress by Taylor factor with M~6.5 for polycrystalline Mg [46]. The theoretical yield stress is found to be 517 MPa from equation 4.17 by substituting appropriate values for Mg such as G=16.7 GPa, =0.35, b=0.321 nm and the critical grain sizes lc=33 nm. The experimental results show that 40h-MMed sample with grain size 33 nm yielded at 205 MPa which is lower than the theoretical value estimated by equation 4.17 indicating a much lower yield stress to activate the deformation. It also inferred that a dislocation can no longer operate below this critical length scale. 99 Processing, physical and tensile properties It can be noticed that yield stress of the 40h-MMed composite sample is comparable to that of the 40h-MMed pure Mg sample. This indicates that after milling 40h, the strengthening effect of grain refinement, solid solution and the reinforcement particles diminished in the composite sample. Moreover, similar ductility was displayed in both composite and pure Mg samples after 40h milling. It can be concluded that there is very high possibility of the same deformation mechanism operating in the 40h-MMed samples of both material systems. When the grain is refined to a critical size, this composite seems to behave as its pure matrix regardless of its composition. If the grain boundary strengthening effect is negligible after certain critical grain size, the minimum stress to activate the dislocation movement may be determined by the lattice frictional stress i. In that case, due to very limited movement of dislocations, very low ductility will be resulted. However, the ~34% elongation of the 40h-MMed samples nullifies this statement. From the experimental investigations, lower YS and absence of work hardening in these nano grain sized samples may imply the onset of a different deformation which is different from coarse-grained counter part. When grain size becomes smaller, volume fraction of grain boundary becomes more significant and hence diffusional deformation process may play an important role at room temperature deformation. Thus the Hall-Petch strengthening phenomenon diminishes while creep related grain boundary weakening process may operate in nano grain size region. The disordered grain boundary can act as a sink for dislocation within the grain and lattice dislocation can travel across and be absorbed by the grain boundary leaving the grains dislocation free and increasing the grain boundary free volume. Assuming the 100 Processing, physical and tensile properties grain size to be d and grain boundary thickness, , the volume fraction of the interface Vf can be approximated by [47]:  d  V f  1    d   (4.18) From equation 4.18, high volume fraction Vf of interface of about 8.6% may be estimated for the 40h-MMed composite sample with d~33 nm and ~1 nm. However, it is calculated to be only 0.0176% for coarse-grain sample with 17µm. A significant increase in volume fraction of the interfaces in the 40h-MMed sample implies a strong possibility of highly localized deformation in the grain boundary region. A number of investigators have suggested that the inverse Hall–Petch relation can be attributed to the increased grain-boundary activity due to grain-boundary sliding and/or diffusional mass transfer via grain-boundary diffusion [48-50]. For example, Masumura et al. [51] suggested that the strength softening with decreasing grain size is due to the competition between conventional dislocation motion and grain-boundary diffusion via Coble creep, which is assumed to be responsible for the roomtemperature plastic deformation of nc materials. However, it is not clear how Coble creep [52], which successfully describes the creep mechanism of coarse-grained polycrystals, can be extended to nc materials with grain sizes of several nanometers. Moreover, the experimental data indicates a mild grain-size dependence of YS in the strength-softening region [53], in contrast to very strong grain-size dependence, as required by Coble creep. 101 Processing, physical and tensile properties Grain boundary diffusivity Dgb is given as  Q  Dgb  D0 exp    RT  (4.19) where D0 is a constant called pre-exponential factor, Q the activation energy for the diffusion process, R the universal gas constant (8.314 J/mole K) and T the absolute temperature. It was suggested that grain boundary diffusion is significant only for deformation in some fine grains at low homologous temperature of T/Tm[...].. .Processing, physical and tensile properties Table 4. 1 Grain sizes (in nm) of powders (P) and extruded samples (E) after different milling durations Composition Mg- 5Al- 1AlN (P) Mg- 5Al- 1AlN (E) Mg (P) Mg (E) * m 0h 24* 17* 24* 25* 10h 44 116 41 183 20h 32 86 31 158 30h 26 42 28 127 40 h 22 33 25 144 Shear deformation during hot extrusion caused grain refinement in the as-blended or 0h-MMed extruded composite. .. Electrical resistivity of Mg- 5Al- 1AlN composite sample and pure Mg samples 4. 3.3 Thermal properties From Table 4. 3 and Fig 4. 8, the same trend of αc and Mg is observed for both composite and pure Mg samples Milling increases the α value with grain refinement accompanying the increasing grain boundary volume after 10h-MM As the CTE of grain boundary is larger than that of crystalline state (2.5 – 5 times... degrees The melting temperatures of MMed samples increase near to Tm of Mg matrix (650°C) after 40 h of milling indicating no apparent influence of second phases and reinforcement on melting with longer milling duration It is interesting to note that milling has no influence on the melting of pure Mg samples exhibiting similar melting temperature of 645 - 646 °C in all samples as shown in Fig 4. 9 (b) 0h 30h... pronounced for pure Mg indicating grain refinement strengthening mechanism dominates in the initial milling of 10h and 20h in the composite samples The presence of AlN particles induces an inhomogeneous deformation pattern and a high dislocation density in the composite matrix, thus leading to higher YS in the composite matrix compared to the unreinforced pure Mg samples During tensile test, a lot of geometrically... processing flaw and structural imperfection  of composite samples increased up to 20h-MM followed by a decrease after 30 and 40 h of MM This might indicate 81 Processing, physical and tensile properties the lesser processing flaws such as porosity, internal cracks, etc and the diminishing dislocation activities in 30 and 40 h-MMed samples with higher degree of grain refinement as evident in Fig 4. 4... 10h 20h 0.60 40 h 0.50 Cp (J/g.K) Cp (J/K.g) 0.60 0 .40 34 0 10h 20h 40 h 0.50 0 .40 0.30 0.30 0.20 0h 30h 360 380 40 0 42 0 44 0 46 0 T (K ) (a) 0.20 34 0 360 380 40 0 42 0 44 0 46 0 T (K ) (b) Figure 4. 10 Cp values of (a) composite and (b) pure Mg samples for different milling durations It can be seen from Fig 4. 10 that the increase in Cp of nanocrystalline samples with temperature is approximately linear as reported... sizes of the as-milled extruded composite and pure Mg samples were in the range of 116-33 nm and 183-127 nm respectively It is clear that retardation of grain growth was effective in composite specimens by second phase particles such as AlN and in- situ formed Al1 2Mg1 7, MgH2, MgAl2O4, and MgO during sintering and extrusion Such stable grain structure with no grain growth up to 550°C has been reported in. .. enhanced ductility of 28% and 34% respectively Table 4. 4 Yield strength and % elongation of Mg- 5Al- 1AlN and Mg samples for different milling durations at 3.33x10 -4 s-1 strain rate Milling duration (h) 0 10 20 30 40 Mg- 5Al- 1AlN YS Elongation (MPa) (%) 219 12 46 5 5 505 9 3 34 28 205 34 YS (MPa) 122 311 256 216 210 Pure Mg Elongation (%) 10 8 10 18 33 Strain hardening behavior can still be observed in 10h-MMed... value of  in 30h- and 40 h-MMed composite samples may indicate diminishing dislocation activities in grains of smaller size 2 In both composite and pure Mg material systems, 40 h-MMed samples exhibit the lowest thermal expansion 3 Although melting temperature of pure Mg is not affected by milling durations, longer milling duration pushed the melting temperature higher up to melting temperature of pure Mg. .. pure Mg after 40 h of milling in composite samples 4 The increase in Cp values of both composite and pure Mg samples is approximately linear Reinforcement particles and second phase particulates might hinder the lattice vibration and lower the Debye phonon frequency This leads to higher Cp in composite samples 5 Both ΔH and ΔS of composite samples are higher than those of pure Mg samples In composite samples, . immiscibility of AlN in Mg. With increasing milling duration, broadening of XRD peaks and declining in peak intensity are observed due to the reduction in grain size and introduction of microstrain during. different milling durations Composition 0h 10h 20h 30h 40 h Mg- 5Al- 1AlN (P) 24 * 44 32 26 22 Mg- 5Al- 1AlN (E) 17 * 116 86 42 33 Mg (P) 24 * 41 31 28 25 Mg (E) 25 * 183 158 127 144 * m . such as AlN and in- situ formed Al 12 Mg 17 , MgH 2 , MgAl 2 O 4 , and MgO during sintering and extrusion. Such stable grain structure with no grain growth up to 550°C has been reported in commercially