Microstructure and mechanical properties of ultrasonically processed Al-7Si alloy / Y2O3 nanocomposite.

The present study aims to investigate the microstructure and mechanical properties of the A356 aluminum metal matrix composite reinforced with Y2O3 particles. The composite is synthesized by adding 1 and 2 vol.% of reinforcement via stir casting assisted by ultrasonic treatment (UT). Microstructural contemplates show improvement in the dispersion of nano Y2O3 particles and a decrease in the porosity level due to the ultrasound aided synthesis. The UT refines the size of the Y2O3 particles as well as helps to improve their dispersion. The secondary dendrite arm spacing of 2 vol.% Y2O3 reinforced samples with 5 min UT are found to be significantly reduced to 12 μm as compared to that of the as-cast A356 alloy. The addition of 2 vol.% of nanoY2O3 has significantly improved the hardness of the A356 alloy from 60 HV to 108 HV. A considerable increment in the YS and TS of the A356 alloy is observed with the increase in the amount of Y2O3 and found to further improve with UT. However, minimal reduction in ductility is observed with the addition of Y2O3 as well as ultrasonic treatment.


Introduction
Aluminum-based metal matrix composites (MMCs) are widely used in aerospace, automotive, mining structural, and military applications due to their low density to high strength ratio, wear resistance, high hardness, elevated temperature resistance, and greater stiffness compared to as-cast aluminum alloy [1][2][3]. To improve the mechanical properties of Al alloys, ceramic particulate reinforced composite systems were developed with the help of advanced processing techniques such as stir casting, ultrasonic-assisted casting, mechanical alloying, and severe plastic deformation, etc. Nevertheless, the ultrasonic-assisted casting method is reported as more efficient among the list since it can synthesize an interfacial clean and homogenized composite melt with added advantages of microstructural refinement and degasification [3]. A study by Raghu et al. [4] reported that ultrasonic-assisted casting is an efficient method to fabricate Aluminum/MgAl 2 O 4 composites with clean interfaces. Another study by Nampoothiri et al. [5] revealed that ultrasonic treatment of A356/TiB 2 composite can induce α-aluminum matrix microstructure refinement. The microstructure revealed the transformation of the dendritic structure into fine globular structure along with modification of Si needles. It was evident that the ultrasonic treatment during the process of solidification is an effective method leading to the formation of globular structures.
Similarly, Jia et al. [6] reported on the formation of globular grain structures by UT in A356 samples. UT was reported to break up the dendritic structures and form globular grain structures, resulting in better mechanical properties. Xuan et al. [7] worked on reinforcing Al 2 O 3 nanoparticles in A356 alloy and dispersed the reinforcement with ultrasonic cavitation processing. The microstructure was reported to refine into globular grain structures from dendritic grain structure on subjecting it to ultrasonic cavitation.
A recent study by R.K. Gupta et al. [8] states that ultrasonic treatment of Al-Cu alloy/Graphite composites can induce a particulate refinement and matrix microstructural refinements. Gupta reported that the combined effect of particulate refinement and microstructural refinement resulted in doubled the tensile properties of composite material when compared to the Al matrix. Gupta added that post stir casting the ultrasonic treatment-induced ~80% porosity reduction compared to normal stir cast Al-Cu alloy / Gr composites. Gupta proposed that ultrasonic cavitation implosion-assisted particle refinement and degasification is the rationale for mechanical property enhancement.
A variety of studies were reported on the advantages of Al-Si alloys for their castability, corrosion, and wear resistance and it made Al-Si alloys an interesting matrix for metal matrix composite researches. A study by Radhika et al. [9] reported the tribological property enhancement of the LM25 matrix by the addition of SiO 2 particles and Arunagiri et al. [10] reported the effect of AlB 2 reinforcement on castability and wear behavior of LM25/ AlB 2 composites. A recent study by Satish et al. [11] revealed that the T6 treatment of Al-7Si/ ZrSiO 4 Composites can exhibit ~4 times better wear resistance compared to matrix alloy. Literature indicates that incorporation of Y 2 O 3 hard ceramic particles by both severe plastic deformation and liquid metallurgy route to the Aluminum matrix can induce mechanical and tribological property enhancement [12][13][14][15][16]. Furthermore, the literature review suggests that the particle size reduction to the nano-regime can improve mechanical properties like yield strength (YS) and ultimate tensile strength (UTS) without affecting the matrix ductility. A study by Shian et al. [17] reported that the agglomeration of nanoparticles in its matrices can happen due to the high surface energy and subsequently it can reduce the mechanical property. But, Su et al. [18] reported that fabrication of Al2024/ nano-Al 2 O 3 composites via mechanical stirring and ultrasonic vibration can improve dispersion of the particles and bonding with the matrix alloy. Yuan et al. [19] successfully produced A356 alloy/nano-SiC p composite using stir casting assisted by ultrasonic vibration and achieved uniform dispersion of nano-SiC p in A356 alloy matrix. Compared to A356 alloy, ultrasonic treated A356 alloy/2 wt.% nano-SiC p composite exhibited significant improvement in mechanical properties like YS, TS, and % El by 62%, 22%, and 24% respectively. Arpanet et al. [20] made-up Al-WC nano-composites using UT with stirring. Significant improvement in the hardness and wear resistance of the UT composite samples was reported to be observed.
Studies reported by Li et al. [21] suggested that the post stir casting UT improves the particle dispersion in ex-situ A356/SiC nanocomposites and the mechanical properties also. Similarly, a study by Khandelwal et al. [22] pointed out that the ultrasonic-assisted casting resulted in noticeable enhancement in the dispersion of nanometric alumina particles in the Mg matrix and also refinement in both particle and grain size when compared to one without ultrasonic treatment. Recent studies by Nampoothiri et al. and Ramani et al. [5,[23][24][25] reported that the UT can refine particle size and grains and improve dispersion in Al/TiB 2 in-situ composites with pure Al. A356 and Al-4.4Cu matrices. Murthy et al. [26] also mention that ultrasonic-assisted casting can improve particle dispersion and thereby mechanical and tribological property enhancement. Further, studies by researchers [27][28][29] also found that UT of aluminum melt such as A356 alloy, Sr modified A356 alloy can remove the dissolved gas inside the to reduce porosity.
Literatures reveals that smaller addition of nanoparticles can enhance the strength of matrix alloy without compromising ductility and also, UT assisted processing can improve particle and microstructural refinement, particle dispersion, and degasification. Furthermore, reports show that, though works have been reported on Al/Y 2 O 3 composites preferably on micro composite or powder metallurgy nanocomposite systems, very limited exploratory works has been reported on an effective scalable method such as ultrasonic-assisted casting for fabrication of Al/Y 2 O 3 nanocomposites. Studies added out that, aluminum alloys reinforced with Y 2 O 3 via powder metallurgy methods can offer substantial property enhancement than other ex-situ composite systems. Though Al/Y 2 O 3 nanocomposites have significant potential in the field of the transport industry, there are very little works has been reported in the scalable processing routes such as casting. Detailed analysis and understanding of structureproperty correlation are essential for the selection and design of composite material for specific applications. This gives research opportunities for developing and/or optimizing a scalable liquid metallurgy process to fabricate Al/Y 2 O 3 nanocomposites. Hence in the present study, an attempt has been made to understand the effect of post-stir casting UT of Al-7Si/Y 2 O 3 composites on its microstructure and mechanical properties.

Experimental Procedure
Aluminum A356 cast alloy and yittria (Y 2 O 3 ) powder (99% pure) were chosen as matrix alloy and reinforcement for the present study. The chemical composition of A356 cast alloy was determined using an optical emission spectroscope (Bruker, Q8 MAGELLAN) and presented in Table 1. The  Fig. 1 and b show the micrograph of the Y 2 O 3 particles confirming the presence of particles with an average size to be ~5 μm.
Approximately 750 g of A356 ingot was charged into the ceramic crucible and melted using a pit-type electrical resistance furnace at a temperature of 750 °C. The required quantity of Y 2 O 3 powder (1 and 2 vol.%) was preheated to 400 °C to remove the volatile impurities. The preheated powders were subsequently added to A356 alloy melt by conventional stir casting method using a zircon coated 3 bladed mild steel stirrer and spatula. To prevent the particle settlement due to the density differences between molten matrix (ρ Al-melt = 2.375 g.cc) and reinforcement (ρ Y2O3 = 5.01 g.cc), an uninterrupted continuous stirring at 600 rpm for 10 min, subsequently transferred into a cast-iron mold of 80 mm length and 20 mm diameter. Thus the prepared stir cast composites are henceforth designated as micro composites in this article.
The stir cast composites were re-melted to a temperature of 730 °C and the melt was subjected to UT for 5 min. A schematic representation of the UT setup is shown in Fig. 2. The working of the magneto restrictive transducer is shown in Fig. 2 and can be briefed as follows; the unit consists of two major parts namely an electromagnetic coil and a ferromagnetic material. Upon supply of the electrical power, the coils in the transducer create an electromagnetic field and the field thus generated induces an atomic level compression in the ferromagnetic material inside the transducer. The compressive stress developed in the ferromagnetic bar induces a compressive strain. The compressive stress developed in the ferromagnetic material can vary according to  the generated electromagnetic field. The cyclic variation in the electromagnetic field generated will vary the compressive strain developed in the ferromagnetic material and it can raise the vibrations [30]. The frequency and amplitudes of the generated vibratory waves can be engineered to the required regime of ultrasonic waves. The vibration generated can be mechanically amplified with the help of a concentrator and transferred to the melt with the help of a sonotrode. A circulating water line is provided to take away the heat generated during the process. In the present work, a magneto restrictive transducer from RELTEC, Russia is used to generate ultrasonic waves with a frequency of 20.4 kHz and a stainless steel (SS304) sonotrode of 40 mm diameter is used to transfer the generated ultrasonic waves at an intensity of 128 W/cm 2 to the melt. To prevent the temperature drop and wall-crystal formation effect, the sonotrode was coated with zircon and preheated to 730 °C before introduction into the melt. Immediately after completion of ultrasonic treatment, the sonotrode was removed and the molten composite was cast into the mould.
Samples for microstructural studies were sliced from the casting and were ground through 240 to 2000 grit papers, polished with alumina paste, and finally etched using Keller's reagent. For etching, Keller's reagent was wiped over the sample surface and allowed to react for 15 s. And thus etched samples were finally cleaned using the ultrasonic cleaner for about 30 s. Microstructural studies were carried out by using polarized light microscopy (Carl Zeiss Axio Scope A1). Transmission Electron Microscopy (TEM) analysis were carried out using JEOL JEM 2100 operating at 200 Kv. The Secondary Dendrite Arm Spacing (SDAS) was measured by detailed analysis of optical microstructures of the samples using 'Image J' image analyzing software.
Theoretical density (ρ t ) of the as-cast alloy and its composites was calculated using the rule of the mixtures and the real density was measured using the Archimedes method. The density analysis was carried out by weighing cylindrical samples of (ϕ 15 × 20 mm) in both air and distilled water. Afterward, the density was estimated using Eq. (1) [31].
where ρ m is the density of the sample, M is mass in air, M w is the mass of sample in water and ρ w is the density of water. For the current experiment conditions, corresponding to a temperature of 30 °C, the density of water was selected as 0.99565 g/cm 3 [32]. Based on the density measurements, the % porosity (P o ) was calculated using Eq. 2 [33].
Hardness test was carried out using a Vickers hardness testing machine loaded with a diamond indenter carrying a load of 1 kg that to be applied on the sample for 15 s (as per ASTM E384 standards). An average of five readings is accounted for as the respective sample hardness. To study the room temperature tensile properties of the composites, the samples were machined as per ASTM E8M standard and the tensile specimen of 6 mm diameter and 30 mm gauge length was tested on a computerized test. Tensile testing was carried out using a computer-controlled universal testing machine (UTM, make: Instron 4467 USA) with dogbone samples of 30 mm gauge length and an initial strain rate of 0.01 mm/min. Three samples were tested for each composition and consistent results have been reported.

Density Analysis
The density and porosity of the UT A356 alloy and its composite are estimated and presented in Table 2 and Fig. 3 respectively. It can be readily observed from Table 2 that the theoretical density of the Al matrix is increased from 2.713 to 2.736 g/cm 3 by the presence of 1 vol.% of Y 2 O 3 particles and further to 2.759 g/cm 3 upon particle addition increment to 2 vol.%. A similar trend can be observed in experimental density also. The porosity analysis of samples depicted in Fig. 3 shows that the addition of 1 vol.% of Y 2 O 3 via stir casting increased the porosity level of samples from 1.05 to ~1.64% and further increment of Y 2 O 3 particles to 2 vol.% resulted in 2.2% of porosity level. The increment in porosity level in composite samples can be attributed to the air entrapment during the stir casting process and fluidity reduction by particle addition [34,35]. However, post stir casting ultrasonically treated samples showed a significant reduction in the porosity revel. The ultrasonically treated 1 The porosity reduction by UT can be attributed to ultrasonic wave-induced degasification mainly by the non-linear effect of wave propagation through the molten metal [29]. The transmission of ultrasonic waves through melt generates pressure waves inside the liquid melt and associated cyclic compression and rarefaction events also. These events lead to the formation of numerous tiny cavitation bubbles inside the melt. The bubbles developed inside the melt can enhance the dissemination of dissolved gas to the bubble or accumulate over the surface of bubbles. On the progression of this event, the size increased bubbles float to the melt surface and explode to release the gases. In addition, there will be a formation of shockwaves due to the cavitation implosion phenomena. The shock waves thus generated can break the particle agglomerated and improve the dispersion and consequently the fluidity. In these manners, the UT can help to mitigate porosity levels.  Fig. 4(a-e). Fig. 4(f) shows the magnified view of a coarse dendritic structure of α-Aluminum and eutectic Si phase in the matrix A356 alloy. A micrograph of the as-cast A356 alloy (Fig. 4a) shows a coarse dendritic structure of α-aluminum and eutectic silicon phase. The magnified image of Fig. 4(a) shown in Fig. 4(f) reveals the presence of acicular needles of eutectic Si with an average size of ~29 μm. The microstructure analysis presented in Fig. 4 shows that the alloy, composite samples without and with UT exhibit a dendritic morphology. It reveals that the addition of Y 2 O 3 particles and UT has a marginal effect on the dendrite morphology. However, the incorporation of particles and UT reduced the size of the dendrites in the composite samples. From the micrographs of the composite samples, it can be observed that the SDAS of samples with UT is significantly reduced compared to the samples without UT. The average value of measured SDAS of the A356 alloy and its composites is estimated and the as-cast A356 alloy has an SDAS of ~29 μm. However, it is observed that 1 vol.% and 2 vol.% of Y 2 O 3 particles added samples have an average SDAS of 18.9 ± 1.8 μm and 17.3 ± 1.7 μm respectively. Nevertheless, the performance of UT on 1 and 2 vol.% composite samples resulted in an SDAS of about 15.8 μm and 12.7 μm respectively. Similar results have been explained by Jayakrishnan et al. [36] In their study on UT of A356/TiB 2 in-situ composites. The presence of Y 2 O 3 particles inside melt can increase the events of nucleation by acting as nucleation sites to attribute the SDAS reduction. Further upon the ultrasonic treatment, the particles can be fragmented during the ultrasonic cavitation implosion and the shock waves generated during the implosion can make the surface of the particle beneficial for nucleation [37]. Thus the increased number density of Y 2 O 3 particles and active nuclei particles can contribute to the further reduction of SDAS in the ultrasonically treated samples. In addition, the formation of agglomerates is evident from the micrographs (Fig. 4b and c) and the presence of agglomerates is marginal in UT samples shown in Fig. 4(d and e). Figure 5a shows the SEM image of A356 alloy and  Fig. 5d. It can be inferred from the SEM micrograph that, the Si needles in the A356/1 vol.% composite (Fig. 5(a)) exist in the acicular morphology with an average needle size of 12 ± 3.9 μm. The addition of 2 vol.% of Y 2 O 3 particles via stir casting reduced the average eutectic Si needle size to 7.1 ± 2.4 μm and UT of 1 and 2 vol.% composites resulted in further reduction of eutectic Si size to 4.9 ± 2.8 and 4.4 ± 3 μm respectively. The presence of particles such as Y 2 O 3 can act as pinning agents for the growth of Si needles and the number density enhancement by UT-assisted particle reduction can further restrict the Si growth. Fig. 5(d) reveals that the agglomerates of Y 2 O 3 particles are broken and also the size of particles was significantly reduced from 5 μm to 500 nm due to UT. It can also be observed from Fig. 5(d) that, the nanoparticles are found to be spherical in shape.

Microstructural Analysis
Jayakrishnan et al. [36] achieved a reduction in the size of TiB 2 particles from micron size to nano-size after 5 min of UT and the results of the present study correlate the same. In addition, the Y 2 O 3 particles in the inset of Fig. 5(d and e) are found to be present in both the eutectic Si phase and primary matrix phase, which confirms that the added nano Y 2 O 3 particles were trapped into the solid solution and are not rejected by the solid-liquid interface throughout solidification. The particles present in the eutectic Si phase restrict the grain growth generation and form a Y 2 O 3 rich eutectic domain along with eutectic silicon. A similar effect was reported by Bouaeshi et al. [37] in their reports on Y 2 O 3 addition of aluminum. Puga et al. [38] also supported the above mechanism in their study on the properties of AlSi 9 Cu 3 alloy. Khalifa et al. [39] explained the formation of the eutectic domain along with silicon and its effect. Yu Pan et al. [40] and Kewei Xie et al. [41] observed significant refinement in grain structure with the addition of TiC to Ti and Al nanoparticle to Al respectively.
No other harmful or unconventional intermetallic phases were observed from the SEM micrographs and it may be due to the thermodynamic stability of nano Y 2 O 3 particles. Rahul Gupta et al. [42] accounted the similar outcomes in their study on creep properties of ultrasonically processed in-situ Al 3 Zr-Al alloy composites. During the cavitation implosion, a localized temperature rise to about ~8000 °C and pressure rise to 2-3 MPa have occurred and this temperature can slowly dissolve the Y 2 O 3 particles to the melt and later can be precipitated [36]. During the course of particle precipitation, the shock waves of implosions can break the growth of the particles to form nanoparticles [5]. In addition, the cavitation implosion shock wave produces some "hammering" effect on particles to break to nano regime. Further, the UT acoustic streaming and cavitation non-linear effect gives efficient stirring to distribute the particles throughout the matrix.
TEM bright-field micrograph of the Y 2 O 3 particle (Fig. 6a) shows the size of the particle to be about 500 nm. Fig. 6a reveals the excellent bonding of the particles with the matrix and is free from interfacial reactions. EDS analysis (Fig.6b) further confirms the presence of Y 2 O 3 particles in the composite. The simulated and the observed Selected Area Diffraction (SAD) pattern of nano Y 2 O 3 particles are shown in Fig. 6(c & d). The SAD pattern indexing also confirms the presence of Y 2 O 3 particles, as the d-spacing is following the Y 2 O 3 particles. Fig. 6e shows the magnified view of the area from the α-Al matrix and it shows the presence of string/worm-like structures in the image. The micrograph in Fig. 6(e) matches well with the standard morphology of dislocations reported in the literature [42] the presence of a large number of dislocations near the grain boundary. These dislocations will aid in significantly improving the mechanical properties of the matrix alloy. Further, direct observation and the threshold filtering analysis [43] done with the help of Image J software (Fig. 6(f)) confirmed the presence of the dislocation densities. Figure 7 shows the average hardness values for the monolithic alloy and composites. A356 alloy reinforced with 1 vol.% and 2 vol.% of Y 2 O 3 particle without UT is about 75 HV and 95 HV receptively. Similarly, the average hardness values for the composites reinforced with 1 vol.% and 2 vol.% of nanoY 2 O 3 particles (with UT) is about 85 HV and 108 HV respectively. It can be inferred that the addition of 2 vol.% of nanoY 2 O 3 has significantly improved the hardness of the A356 alloy from 60 HV to 108 HV. It can be seen that the hardness sample increases with UT and the synergistic effect of Y 2 O 3 particles and UT significantly increases the hardness of A356 alloy by about 80%. The presence of high dislocation density in nanocomposites are evident from the TEM micrograph shown in Fig. 6(e) and the combined effect of porosity reduction, SDAS reduction, size reduction of Y 2 O 3 particles to nano regime, and associated dislocation density formations and or enhancements can jointly contribute to the hardness increment. The presence of particles can enhance the hardness through dislocation bypass and interaction effects and also by load transfer effect.

Tensile Properties
The room temperature tensile behavior of A356/Y 2 O 3 composites and monolithic alloy were compared in Fig. 8. It can be noted that the average yield strength (YS), tensile strength (TS), and elongation at break (% El) of 79 MPa, 138 MPa, and 4.94% respectively. 1 vol.% Y 2 O 3 particles added composites exhibited an increase in YS and UTS to 86 MPa and 138 MPa respectively as shown in Fig. 8. Further, 2 vol.% The tensile properties of the samples reinforced with 2 vol.% nano Y 2 O 3 particle added composite with 5 min UT exhibited a YS of 109 MPa and UTS 167 MPa with % of improvement by 27% and 26% respectively. The ultrasonic cavitation treatment significantly enhanced the YS and UTS of the composite. It can be observed that the UT improves the % of elongation and the elongation of 1 vol.% nano Y 2 O 3 particle added composite with UT is higher than that of 2 vol.% nano Y 2 O 3 particle added composite with UT. Generally, composites reinforced with micrometer-sized ceramic particles will reduce the ductility of the matrix due to their stress raiser effect, whereas the nanoY 2 O 3 particle reinforced composites improved the tensile properties by retaining the ductility. Similar results were reported by Yuan et al. [34] for A356 alloy/nano SiC particles reinforced composites prepared by UT.
Significant improvement in the mechanical properties is due to the existence of uniformly dispersed nanoY 2 O 3 particles and is probably contributed to the various strengthening mechanisms. The dominant mechanisms were observed in the present study. The large difference in Coefficient of Thermal Expansion (CTE) between the reinforcement Y 2 O 3 (8.1 × 10 −6 /K) and matrix (25 × 10 −6 /K) may lead to the development of stresses during cooling resulting in the formation of a large number of dislocations around the particles as shown in Fig. 5e.
Orowan strengthening mechanism is one of the dominant strengthening effects and mainly depends on the reinforcement particle size, bond strength with the matrix, and interparticle spacing. Under tensile load, the softer matrix alloy deforms more plastically compared to the harder nano Y 2 O 3 particles. During plastic deformation of the composite samples, closely spaced nano Y 2 O 3 particles will become a barrier to the dislocation motion thus hindering their motion. As a consequence, dislocations will bend and form a loop in the region of Y 2 O 3 particles. Consecutive dislocation incessantly forms a loop in the region of Y 2 O 3 which brings on back stress and obstructs the dislocation movement further.
In this manner, the Y 2 O 3 nanoparticle facilitates the strengthening of the A356 alloy matrix. Compared to the stir cast composites (without UT), inter-particle spacing in the nanocomposites (with UT) will be very less and thus offers more resistance to the dislocation movement and thus further strengthening the matrix. According to Hall-Petch theory, the fine grains with more grain boundaries will significantly improve the strength of the matrix [44]. In the present study, the combined effect of Orowan looping, CTE mismatch, load transfer effect, and SDAS refinement can contribute to mechanical property enhancement. The effect of each phenomenon can be quantified using Eqs. 3-8. For the calculation purpose, the Hall-Petch term is calculated using SDAS data since a study by Ehsan et al. [45] reported that, for Al-Si hypoeutectic alloys, secondary arm spacing fits better than grain size.
The theoretical yield strength (σ y ) of A356/Y 2 O 3 composites can be written as: Fig. 7 Microhardness values of A356 alloy and its composites without UT and with UT. where, k is the Hall-Petch constant, d grain size or SDAS, σ m is the matrix yield strength, σ oro is the Orowan strengthening, σ CTE is the contribution from stress developed by CTE mismatch between particles and matrix, σ Geo is the stress contribution due to strain gradient associated with geometrically necessary distributions required to accommodate plastic deformation mismatch between matrix and reinforcement particles and σ Load is the influence of load transfer effect between particles and matrix [36].
The Orowan strengthening can be calculated as [24]: where m is Taylor's factor, G is the shear modulus, b is the Burgers vector, ϕ is the particle size and λ is the planar interparticle separation, ∕ √ V p , where V p is the volume fraction of particles.
The strength contribution from CTE mismatch can be calculated as [36]: where η = 1, and ρ is the dislocation density, Where, Δα ∆αis the difference in coefficient of thermal expansion f matrix alloy and reinforcement ΔT∆T is the difference in testing and processing temperature.
The strength increments by geometrically necessary dislocations are [36].
Where ββ is a geometric factor with a numerical value of 0.2 and ε is the plastic strain of the matrix.
And the contribution of load transfer effect is calculated as [36], Where s is the aspect ratio of particles.
The values of parameters used that for theoretical calculations of yield strength of A356/Y 2 O 3 composites are given in Table 3.
The theoretically calculated and experimental tensile data is consolidated in Table 4 for ready references.
It can be inferred from Table 4 that, the yield strength of composite without UT has anomaly with theoretical values than the monolithic alloy and UT composites. This anomalous behavior in composites without UT can be attributed to inhomogeneity in the samples. On the other hand, the experimental YS of other composites is closely matching with the theoretical value. However, the small error in the theoretically calculated YS and experimentally measured YS can be rationalized by the unavoidable errors in machine compliance corrections, leftover porosities, and small variations in elastic constants of the samples.  The percentage contribution from each strengthening model is analyzed and plotted in Fig. 9. It reveals that the Hall-Petch mode is the main strengthening mechanism when the particle size is in micron regime and is followed by CTE mismatch strengthening. However, as the particle size is reduced to 500 nm, the strengthening contribution from CTE mismatch and Hall-Petch strengthening become nearly equal. The contribution of Orowan strengthening and geometrically necessary dislocations also considerably increased when the particle size is reduced to 500 nm. It can be inferred from Fig. 9 that, in A356/Y 2 O 3 micron composites, grain boundary strengthening is the primary mechanism and in nanocomposites, CTE mismatch is the activated mode of strengthening.
Fractography examinations of tensile-tested samples (for both stir cast and UST) have been carried out to understand the effect of the Y 2 O 3 particles in the fracture mechanism of the A356/Y 2 O 3 composite samples. Fig. 10a and Fig. 10(b-e) show the SEM micrographs of A356 alloy and Y 2 O 3 particle added composites respectively after the tensile test. SEM image of the A356 alloy Fig. 10a shows a ductile fracture, whereas the composites show a cleavage fracture. From the surface morphology of stir cast samples and UT samples (Fig. 10b-e), it can be observed that there are sufficient tear ridges on the fracture surface, demonstrating the quasi-cleavage fracture as the main fracture mechanism. Apart from this, the UT samples show the presence of tiny dimples on the fracture surface leading to increased ductility as compared to the stir cast samples.
Good interfacial bonding between the Y 2 O 3 particles and the A356 alloy prompts effective load transfer from the Y 2 O 3 particles to the A356 alloy. This in turn reduces the crack propagation leading to a major improvement in the tensile properties of the composites with a reduction in ductility [46][47][48][49][50].
The present study infers that post stir casting ultrasonic treatment of Al/Y 2 O 3 micron composite is a novel and facile method to synthesize Al/Y 2 O 3 nanocomposites. This process can refine micro-sized Y 2 O 3 particles into nano regimes such as 500 nm and disperse the particles through the matrix. This process is scalable to mass production too. As a limitation of the present study, the authors propose that; since ceramic particles below 100 nm can significantly improve the mechanical properties with the addition of less volume fraction of particles, further studies are required to extend Y 2 O 3 size refinement to the prescribed regime below 100 nm.

Conclusions
From the present study, it can be inferred that the post stir casting UT can reduce the particle size from micron to nano regime. The addition of Y 2 O 3 particles in both sizes is found to induce a significant reduction in SDAS. However, more reduction can be observed in A356/2 vol.% of Y 2 O 3 nanocomposites. Increased particle number density and ultrasonic-assisted activation of nuclei particles are suggested as the rationale for the SDAS refinement of nanocomposite samples. However, a detailed study is required to understand the effect of Y 2 O 3 and UT on dendritic modification and SDAS refinement.
The ultrasonic treatment of stir cast composite is effective in melt degasification and associated improvement in the density of cast product. The combined effect of reinforcement particle refinement, SDAS refinement, and degasification contributed significantly towards the mechanical properties of composites. The theoretical model-based analysis on strengthening mechanism revealed that in A356/Y 2 O 3 micron composites, grain boundary strengthening is the governing one, and while the particle size is reduced to 500 nm, CTE mismatch strengthening becomes the predominant mechanism.