The evaluation of microstructure, grain boundary character and micro texture of [Al/Si3N4/Al2O3] P nanocomposites fabricated through PM route and its influence on compressive and three-body wear properties

The compressive properties and 3 body wear characteristics of powder metallurgical (PM) processed [Al/Si3N4/Al2O3] P Nanocomposites with single and combined reinforcement of Al2O3 and Si3N4 reinforcing particles having different compositions (1%, 2% and 3%) were studied and evolution of microstructure, grain boundary character and micro texture of fabricated [Al/Si3N4/Al2O3] P Nanocomposites was investigated through EBSD in the present research work. The fraction of high angle boundaries (HAGBs) were observed more in combined reinforcing samples of Al2O3 and Si3N4 whereas a single reinforcing sample of Al2O3 and Si3N4 showed fewer HAGBs. Micro texture results showed the strong textures components near to {112}〈111 〉 Cu and {110}〈111 for pure sintered Al sample P and mixed reinforcement composites (M1, M2 and M3) > P whereas for single reinforcing sample showed weak textures near to transverse direction. Out of all fabricated composites, 2% mixed Al2O3 and Si3N4 reinforced composite revealed the maximum ultimate compressive strength (209.98 MPa) and least wear rate (0.1 mm3/min mm3/N-cm for 1 kg load and 3.5 mm3/N-cm for 2 kg load) attributing formation of nanocluster causing grain boundary pinning effect. The dominant failure mechanism for all samples was also detected and found to be a mixed-mode ductile failure mechanism for 2% mixed Al2O3 and Si3N4 reinforcement composite while other sample failed through ductile as well as mixed-mode mechanisms.


Introduction
Aluminium-based metal matrix composite (AMMC) extensively used in automobile parts, aircraft structure and electronic devices due to their enhanced properties such as high specific strength, hardness, fatigue and fracture toughness compared to monolithic materials [1]. Out of all AMMCs, metal matrix composites based on nanoparticle reinforcement has emerged as a promising material and found most attention owing to the presence of hard and brittle nano reinforcement which provides better tribological properties, creep characteristics and cyclic fatigue/fracture response [2]. The combined advantage of matrix material and reinforcing particle may also offer the synergistic enhancement in material properties which is unachievable in unreinforced material.
Various techniques have been proposed to fabricate nanocomposites with improved mechanical and tribological properties and can be categorised into two ways: liquid state processing and solid-state processing. The liquid state processing techniques involve; squeeze/stir casting, pressure filtration, selective laser melting and ultrasonic-assisted casting [3,4]. These processes typically involve the integration of nano reinforcing

Material and methods
In this research work, very fine laboratory-grade aluminium powder of 600 meshes is used for matrix material with average particle size (APS) up to 20 microns. The reinforcing materials to fabricate the nanocomposites are nanoscale particles of silicon nitride (Si 3 N 4 ) and alumina (Al 2 O 3 ). Both as received nanoparticles are having a purity of 99.9%. To develop the nanocomposites, the powder metallurgy (PM) route was used under the compaction pressure of 12.06 MPa. The matrix and reinforced particles were premixed to develop green compacts according to the various wt% (1, 2 and 3 wt %) as shown and abbreviated in table 1.
The microstructural analysis of developed nanocomposites was done using Transmission Electron Microscopy (TEM) and Electron Backscatter Diffraction (EBSD). TEM investigations are carried out through Philips CM 12 transmission electron microscope operating at 120 V. TEM samples for pure and fabricated composites were made by mechanical polishing and grinding the sample to 100 μm with the help of emery papers and polishing machine followed by thinning the sample by twin jet electro polishing with a solution of 20% sulphuric acid and 80% methanol at a temperature of −20°. Microstructural characterization using EBSD was carried out with the help of FEI QUANTA 400 FEG SEM. The samples were mechanically ground and polished through diamond paste with a fine sized colloidal silica suspension for sufficient time. EBSD data measurement using an orientation map was performed by taking a minimum step size of .03. The TSL OIM software was used for EBSD analysis. Grain size measurement and orientation of grains were characterized through an inverse pole figure (IPF) map. Texture information and grain size are related to the different colour associated with the IPF map. For the identification of grain boundary misorientation angle, a boundary misorientation histogram was also plotted. If the adjacent grains have a misorientation angle (θ) of 2 0 (θ)10 0 then those grains were considered as low angle grain boundaries (LAGBs) whereas the misorientation angle (θ)10 0 was treated as high angle grain boundaries (HAGBs).
The phase identification of all fabricated nanocomposites was done using x-Ray diffraction (XRD) subjected to the diffraction angle 2θ (20°to 80°). To analyse the monotonic behaviour of prepared nanocomposite samples, the compressive test was carried out for all sample conditions. The test samples were prepared as per ASTM standard E9 [13]. To analyse the tribological behaviour of the fabricated nanocomposite sample, three body wear test was performed on each sample under varying loading condition (1 kg and 2 kg) for a constant time of 10 min. To confirm the mechanism of wear the scanning electron microscopy (SEM) was done post experimentation.   Figure 1 shows the XRD graphs for all nanocomposites. The well-structured aluminium phase is observed by XRD characteristics in all fabricated samples as evidenced by sharp diffraction peaks. For 1 wt% Al2O3 (A 1 ) reinforced sample condition very small size peaks are clearly visible which confirms the presence of A1. With an increasing percentage of Al 2 O 3 reinforcement from 1 wt% to 3 wt% the gradual rise in peak intensities were observed (figure 1). Similarly very small size peak of sample condition S1, i.e. 1 wt% reinforcement of Si3N4 was noticed in XRD graph (figure 1). As the percentage reinforcement of Si 3 N 4 increases from 1 wt% to 2 and 3 wt%, peak intensity increases correspondingly as depicted from figure 1. Moreover for the sample conditions M1, M2 and M3 both Al 2 O 3 and Si 3 N 4 phases were observed, however with increasing percentage of these phases, gradual rise in the peak intensity is noticed from M1 to M3 sample conditions.

Electron back scattered diffraction (EBSD)
Microstructure charecterisation for pure and fabricated nanocomposites were performed by using Electron back scattered diffraction (EBSD) technique in the present work. The microstructural details and grain size distribution of pure sintered Al sample (P) and fabricated nano composites as achieved from EBSD analysis, are displayed in figures (3)(4)(5)(6)(7)(8)(9)(10)(11)(12).   figure 5(c) which clearly confirms that large fraction of grains in sample are sizes lesser than 1 μm. Bounadry orientation histogram as seen in figure 5(d) for this sample composition represents the large fraction of LAGBs and slight increase in fraction of HAGBs with respect to A 1 sample composition which indicates the better refinement and homoginity in the Si 3 N 4 reinforced composite.
EBSD results for 1% mixed Al 2 O 3 and Si 3 N 4 i.e. M 1 sample composition are presented in figures 6(a)-(d). As observed in figures 6(a)-(b), which are IPF and IQ maps of M 1 sample, the siginificant changes in the microstructure was noticed here. The microstructure consisted of wide range of grain size ranging from 100 nm to 12 μm with an average grain size if 780 nm. In addition,the random orientation of grains are also observed for M 1 sample composition as noticed from IPF and IQ maps. This might be due to inhomogenity during sintering and mixing process of two indifferent reinforcing particles in the present work. IPF and IQ maps also reveals that number of fractions of grains which are sizes lesser than 1 μm are lower as compared to single reinforcement of Al 2 O 3 and Si 3 N 4 . Boundary misorientation histogram for M 1 sample composition is shown in figure 6(d). The slight decrease in fraction of LAGBs and drastic increase in HAGBs compared to single reinforcement samples (i.e. A 1 and S 1 ) can clearly be observed in this figure which validates the lower deformation strain during the processing of nanocomposite through PM process.
EBSD results for 2% Al 2 O 3 (A 2 ) sample composition are shown in figures 7(a)-(d). As seen in figures 7(a)-(b), IPF and IQ maps of this condition was almost similar as in case of A 1 sample composition, however fraction of black point (unindexed point) was observed to be less compared to 1% Al 2 O 3 reinforcement (A 1 ). The reduced fraction of black spot or unindexed spots also confirms that with increasing wt % of Al 2 O 3 particles porosity level in sample decreases gradually. Grain size distribution chart for A 2 sample composition is presented in figure 7(c). In this case microstructure consisted of grain sizes ranging from 100 nm to 140 nm with an average grain size of 119 nm as predicted from grain size distribution chart (figure 7 (c)). In addition, from IPF, IQ and grain size distribution chart, it is also noticed that more than 85% grain attains the grain size below than 1μm which clearly indicates the more homogeneous and finer microstructure for A 2 sample. Boundary misorientation histogram for A 2 sample composition is shown in figure 7(d). High fraction of LAGBs and lower   EBSD results for mixed 2% Al 2 O 3 and Si 3 N 4 (M 2 ) sample composition are shown in figures 9 (a)-(d) . As seen in figures 9 (a)-(b) for IPF and IQ map, microstructure consisted of various ranges of grain size and catagorised in three different zone. One zone consisted of very fine nanometer level grain structure having grain size in range of 40 to 100 nm. In another zone grain size lies in the range of 100 nm to 700 nm,while in third zone more than 700 nm to 5 μm grain sizes are observed. IPF and IQ map also gives the glimpse of recrystallization of sample in this case. Grain size distribution chart for this sample composition is shown in figure 9(c) which clearly indicates the wide grain size distribution for this condition. The quantitative data for the fraction of high and low angle grain boundaries for this sample composition is represented by boundary misorientation histogram in figure 9(d). The fraction of LAGBs slightly decreases and fraction of HAGBs has noticeably been increased as observed in figure which clearly validates that recrystallization process has occurred in this sample composition.
EBSD results for sample A 3 , S 3 and M 3 are shown in figures (10-12). On close observing these results it can be mentioned that on single reinforcement of Al 2 O 3 and Si 3 N 4 up to 3%, microstructure remains in homogeneous state and on hybrid reinforcement of Al 2 O 3 and Si 3 N 4 up to 3% wide range of grain sizes were obsreved as seen in figures (10)(11)(12). The average grain size for A 3 , S 3 and M 3 were noticed as 167 nm, 225 nm and 810 nm . The boundary misorientation histogram for A 3 and S 3 condition represents high fraction of LAGBs while low fraction of HAGBs, on the other hand M 3 sample composition shows just reverse of it i.e. low fraction of LAGBs and high fraction of HAGBs.        from IPF map corresponding to this sample condition ( figure 3(d)) also validates the presence of recrystallised texture in this case [15][16][17]. In addition maximum intensity of recrystallized texture is 9.4 as seen from from figure 13(a). These textures also represents that prefered texture componants in the sample are near to {112} 〈111〉Cu and {110}〈111〉P for pure sintered Al (P) sample. It is believed that copper componant may be produced during premixing and compacting process while phosphorous componant is evolved during sintering process.
Pole figures (PFs) for A 1, S 1 and M 1 sample compositions are shown in Figure 13 (b) and figures 14(a), (b). PFs for single 1% reinforcement of Al 2 O 3 and Si 3 N 4 samples i.e. A 1 and S 1 shows unrecrystallised microstructure with texture near to transverse direction (TD). The presence of high fraction of LAGBs as seen in EBSD IPF map also confirms the same. In addition intensity of preferred texture componant for both samples A 1 and S 1 siginficantly reduced to 1.5 as noticed from figure 13(b) and figure 14 (a). The drastic drop in texture intensity supports the claim of random orientation of crystallographic planes in the present work [15][16][17]. The significant change in the texture devlopement is observed for the 1% mixed reinforcement of Al 2 O 3 and Si 3 N 4 (M 1 ) sample composition as seen in figure 14 (b). The PFs for this sample composition shows recrystallised microstructure with strong texture near to transverse direction (TD). The observed preffered texture componant in the sample are near to {112}〈111〉Cu and {110}〈111〉P as in case of pure sintered Al (P) sample. On comparing the texture intensity with respect A 1 and S 1 samples, it is seen that intensitry reaches from 1.5 to 2.7 for M 1 composition. The improvement in texture intensity for M 1 composition clearly explains that mixed reinforcement has certainly preffered orientation in TD. Moreover high fraction of HAGBs as observed in EBSD IPF map validate the trace of recrysallization during PM processing of this sample.  figure 15 (a). As seen in figure 15 (a) , A 2 sample has a weak texture in TD as in case of A 1 sample ( figure 13 (b)). EBSD IPF map for this sample also does not show any kind of dynamic recrysallisation process as observed in figure 7 (a). The complete microstructure pertains nano sized grains with large number of LAGBs as noticed from grain size distribution chart ( figure 7 (c)). It may be mentioned that with increasing % reinforcement of Al 2 O 3 recrystallisation process could not occur resulting weakning of texture. In the present work, texture intensity level for A 2 sample was observed as 1.5 same as in case of A 1 sample. Similarly PFs for S 2 sample with 2% Si 3 N 4 addition are dislpayed in figure 15 (b). It is noted that with increasing % of Si 3 N 4 dynamic recrystallisation is belived to occur as noted in {001}, {110} and {111} pole figures and EBSD IPF map (figures 8 (a)). High fraction of HAGBs in this case also reported the same as seen fron grain size distribution chart ( figure 8 (c)). On comparing the texture intensities with single reinforcement of Al 2 O 3 and Si 3 N 4 (A 1, S 1 ) , it is found that texture intensity rises from 1.5 to 1.9 which indicates that S 2 sample has slight preffered orientation near to {112} 〈111〉Cu and {110}〈111 P. PFs for M 2 sample with 2% mixed compositions of Al 2 O 3 and Si 3 N 4 are shown in figure 16 (a). PFs of M 2 sample clearly reveals the substantial change in preffered orientation and texture development. It is also noted that dynamic recrystallization process has been accelerated with % increase of mixed reinforcement of Al 2 O 3 and Si 3 N 4 up to 2% as observed in figure 16 (a). More fraction of HAGBs (figure 9 (d)) corresponding to this sample composition also validates the same. On close revealing the texture intensity of M 2 sample, the texture intensity reaches to 3.7 indicating strong texture towards TD direction.
Similary {001}, {110} and {111} pole figures (PFs) for A 3, S 3 and M 3 sample are shown in figure 16(b) and figures 17 (a)-(b) respectively. The observed trend for texture development for all these samples were noticed same as in case of A 1 , S 1 , M 1 and A 2 , S 2 , M 2 in the present work. In this case single reinforement samples of 3% Al 2 O 3 and Si 3 N 4 has shown slightly improved texture intensities of 2.1 and 2.3 with respect to single reinforcement of 1% and 2% Al 2 O 3 and Si 3 N 4. However with 3% mixed Al 2 O 3 and Si 3 N 4 sample drastic increase in texture intensity 4.9 was observed which substantiate the strong texture of M 3 sample in TD as compared to rest of fabricated nanocomposite samples. It is also noticed that with increasing % of mixed reiforcement up to 3% recrysallisation mechanism has prominience effect on the microstructure as seen in IPF and grain size distribution chart (figures 12 (a)-(c)).

Mechanical behaviour 3.3.1. Hardness
Hardness is very useful property that provides the information about the strength and tribological behaviour of processed material. In general various factors such as particle shape/size, volume fraction, dispersal of reinforcing phase, grain boundary nature, texture involved and preparation method may affect the hardness of fabricated composite as reported in literature [8,18]. Figure 18 represents the average hardness versus compositions graph for all fabricated nanocomposites. Hardness of pure sintered Al sample (P) has a value of 48 HV and found to have lowest hardness values with respect to rest of samples. The lower hardness of pure Al sample may be correlated with EBSD microstructure in section 3.2.2, where coarser grains with average grain size of 40 μm along with high fraction of HAGBs were seen. Coarser grain and high fraction of HAGBs might be the reason for lower hardness of pure sintered Al sample in the present work [19]. The sample A 1 in figure 18   shows the hardness of 53 HV which is 10.4% increment in hardness value as compared to pure sintered Al (P) sample. TEM microstructure ( figure 2(a)) corresponding to this sample condition has shown fine spherical size nano reinforcement of Al 2 O 3, which is discussed in detail in section 3.2.1. This nano size reinforcement may effectively pin the grain boundary and might become strong obstacle for dislocation motion as stated in various literatures causing higher hardness of the sample [19,20]. In addition, nano sized grains having average grain size 127 nm and presence of more fraction of LAGBs as discussed in EBSD section 3.2.2 may also provide the additional obstruction to dislocation movement resulting higher hardness of A 1 sample. Hardness values for S 1 and M 1 samples are observed as 50 and 51 HV respectively and can be seen in figure 18. It is seen that there is not major difference in hardness of S 1 and M 1 samples, however the observed hardness values are (50 and 51 HV) lesser with respect to A 1 sample. The drop in hardness of S 1 and M 1 sample may be attributed to increase in fraction of HAGBs and slight coarsening of grains as observed and discussed in EBSD (figures (5), (6)) IPF, IQ and grain distribution histogram. The lower hardness values of S 1 and M 1 sample can also be correlated with TEM microstructure as described in section 3.2.1. It is reported that size and fraction of reinforcement, second phase particles greatly influence the mechanical properties. A fine reinforcement/second phase particle with large volume fraction imparts greater strengthening effect rather than coarser reinforcing particles. Similar phenomenon might be the reason for the low hardness of S 1 and M 1 sample as described in TEM microstructure corresponding to this sample composition. Figure 18 clearly reveals that 2 wt% mixed reinforced nanocomposite i.e. M 2 sample composition is having maximum hardness of 56 HV out of A 2 , S 2 and M 2 sample composition whereas the A 2 and S 2 sample compositions shows the hardness of 53 and 50 HV respectively. High  hardness of M 2 sample may be due to the Nano metric microstructure observed in EBSD IPF and IQ map as noticed in figures 9(a)-(c). Further TEM microstructure for M 2 sample as seen in figure 2(h) also confirms and validates the higher hardness due to uniform clustering of nano reinforcement. It is reported that uniform clustering of nano reinforcement act as an obstacle for the dislocation motion and impose strengthening in the material [21] due to which M 2 sample has shown higher hardness value in the present work. Similarly lower hardness values of A 3 (50 HV), S 3 (50 HV) and M 3 (51 HV) can be explained and correlated with TEM and EBSD microstructure where more fraction of HAGBs, recrystallization effect and coarser reinforcing particles were observed. Figure 19 shows the variation in load with respect to extension for compression test of all fabricated nanocomposite samples along with pure aluminium sample. The responses with respect to compositions provide the moderation in compression property. From figure 19 pure sintered samples have an ultimate compressive strength (UCS) of 81.82 MPa and failure strain 77.54%. This may be due to the coarser grain and high fraction of high angle grain boundaries which provide good properties under compressive loading. Fractured SEM images of pure sintered Al sample is represented in figure 20(a). Post fracture surface morphology of pure sintered aluminium sample (P) clearly shows the large sized dimples ascribing ductile fracture behaviour under compressive loading resulting lesser compressive strength [22]. On addition of 1 wt% Al 2 O 3 (A 1 ) and 1 wt% Si 3 N 4 (S 1 ), UCS becomes 207.86 MPa and 183.94 MPa whereas failure strain reaches to 72.39%and 75.08% respectively. Post fracture surface morphologies of these samples are shown in figures 20 (b)-(c). A 1 and S 1 sample fracture surface consists of finer ductile dimples, river pattern and brittle facets as evident from figures 20(b) and (c). However dimpled features are more to be observed in S 1 sample. Such type of failure mode may be considered as mix mode failure as reported in various literatures [14,22]. In addition dimpled like features in fractured surface indicates the material's plasticity and can be interrelated with the observed fracture strain (75.08%) of the sample. Higher failure strain (72.39%) of A 1 sample might be due to the ductle dimples in fractured surface which accumulate higher strain hardening effect during compressive testing with respect to S 1 sample composition in this work. The UCS and corresponding failure strain of M 1 sample is shown in figure 19. The M 1 sample shows the UCS of 183 MPa while very less failure strain of 7.5% as evident Morover fractured compressive surface of M 1 sample reveal micro crack at an angle of 45°to the compressive test axis attributing the shear mode fracture in the present case. It is reported that shear mode fracture predominantly occurs due to the hetrogeneous deformation and decereased rate of work hardening [23]. Similar behaviour might be the reason for the shear failure and observed lesser failure strain (7.5) of the M 1 sample due to poor debonding of 1 wt% mixed Al 2 O 3 and Si 3 N 4 nano reinforcing particles with matrix.

Compressive properties and post fracture surface morphologyof fabricated composite
Variation of UCS and failure strain for A 2 , S 2 and M 2 sample compositions are shown in figure 19. A 2 , S 2 and M 2 sample have the UCS values of 199.17 MPa, 145 MPa and 209.98 MPa respectively while their corresponding failure strain was noticed as 39.34, 34.98 and 69.77. Post fractured surface morphologies of A 2 sample compositions as seen in figure 21 (a) clearly depicts the presence of fine dimples along with brittle facets for A 2 sample. However the fraction of dimpled covered region is more compared to brittle region substantiating mix mode ductile failure mechanism. On the other hand fractured surface of S 2 sample perturbed with brittle facets and micro cracks at an angle of 45 degree to the test axis as evidenced from figure 21(b). Therefore shear mode fracture mechanism can be noticed for sample composition S 2 in the present work. The observed failure strain for A 2 and S 2 validates this indifferent failure behaviour of these samples as seen from figures 21(a), (b). It may be mentioned that Al 2 O 3 and Si 3 N 4 particles fractured in the dimpled wall and brittle zone may provide the crack initiation cite causing mix mode and shear fracture mechanism for A 2 and S 2 sample composition respectively. Figure 21(c) shows the fracture surface morphology for M 2 sample composition after compression test. Fracture surface of M 2 sample entirely covered with dimpled like features with little amount of brittle zone as seen in this figure. The work hardening ability of this sample composition is believed to be increased significantly with respect to A 2 and S 2 sample composition and can be confirmed from corresponding failure strain. In addition 2 wt% mixed Al 2 O 3 and Si 3 N 4 has shown better interfacial bonding, agglomeration and clustering with matrix in the present work which promotes void formation causing more fraction of dimpled features in fractured surface. It is also reported that excellent interfacial bonding between reinforcing particles and matrix improves the effective load transfer between them and encourages the work hardening behavior [24]. Similar phenomenon might be the reason for improved failure strain of M 2 sample composition in this work. Higher failure strain of A 3 and M 3 samples may be due to ductile and mixed mode failure behaviour of these samples which accumulate sufficient plasticity before the fracture of these samples.

Tribological behaviour (three body wear)
Tribological behaviours of fabricated nanocomposites were observed by three body wear test. In general, sand particles are used as hard protuberances which are forced against the surface of fabricated nanocomposites samples and produces abrasive wear in the sample. Figure 23 shows the specific wear rate with respect to all compositions of pure and fabricated nanocomposites (P to M 3 ). The specific wear rates were studied with varying load (1kg and 2 kg) while other factors are kept constant (travelling distances 1.38 Km and sliding speed 2.3 m s −1 ). Figure 23 clearly reveals that pure aluminium sample has shown the maximum wear under both loading conditions whereas the 2 wt% mixed reinforced nanocomposite (M 2 ) reflects the minimum wear rate (0.1 mm 3 /min mm 3 /N-cm for 1 kg load and 3.5 mm 3 /N-cm for 2kg load) among all fabricated nanocomposite samples for both loading conditions.

Wear surface morphology
Worn surface morphologies of pure sintered Al sample (P) and all fabricated composites are shown in figures (24)- (26). The hardness of fabricated nanocomposites provides property moderation in the form of wear resistance as discussed in section 3.3.1. Hardness of the material is inversely proportional to the wear property as per the archard wear equation reported in the literature [25,26] i.e. greater the hardness lower the wear and vice versa. It has been observed in the present research work that the wear rate is inversely proportional to the hardness of nanocomposite. Worn surfaces of pure (P) and all 1 wt% reinforcement samples i.e. A 1 , S 1 and M 1 are displayed in figure 24.It is depicted from figure 24(a) that pure aluminium sample have maximum wear rate due to lesser hardness and maximum ductility [26]. The wear surface morphology reveals the wear debris in the form of chips due to cutting action. The wear mechanism is totally plastic flow which results large size chips ploughed by harder asperity. In the category of 1 wt% reinforcement nanocomposites all three compositions (A 1 , S 1 and M 1 ) shows brittle nature (figures 24(b)-(d)). Crack propagation has been clearly seen on the surface of all 1 wt% reinforcement nanocomposite samples as observed in figure. Sample S 1 have shown maximum wear  rate whereas A 1 sample is having least wear rate among all 1% reinforcement composites as seen in figure 23. This can be confirmed through respective worn surfaces of A 1 , S 1 and M 1 samples in figures 24(b)-(d) where higher surface damage effects along with pull off of reinforcing particles and majority of cracks were noticed for S 1 sample composition. EDX images has also been shown as an inset for pure and all 1% reinforcement composites in figures 24(b)-(d). From EDX of all 1% reinforcement samples (A 1 , S 1 and M 1 ), it is clearly observed that crack nucleation begins at the respective reinforcing particles [27]. It is believed that when the hard asperity slides against the nanocomposite sample then the brittle nanocomposite could not accommodate the large amount of strain energy resulting development of sub surface crack [28]. Those sub-surface cracks propagate and reaches to the upper surface of material and hence material removal occurs.
In case of 2 wt% reinforced nanocomposites (A 2 , S 2 and M 2 ) sample composition, the sample M 2 has shown least wear rate among all 2 wt% reinforced sample conditions, however S 2 sample composition is having maximum wear rate as displayed in figure 23.The observed wear behaviour of 2 wt% reinforced samples can be correlated with worn surface morphologies of respective samples in figures 25(a)-(c) Worn surface of sample composition A 2 clearly depicts the wear debris and micro crack in figure 25(a). In addition, particle pull off can clearly be evidenced from the worn surface which can be validated through EDX image shown as an inset in figure 25(a). Similar features were also noticed for sample composition S 2 along with large size chips as depicted in worn surface of sample S 2 in figure 25(b). However the surface damage effects were observed more in sample composition S 2 as compare to A 2 Sample composition. Worn surface of M 2 sample composition is shown in figure 25(c). Worn surface of M 2 sample composition have very few wear debris and lesser surface intrusions. Moreover wear particle plugging takes place through plastic flow wear mechanism as observed in figure 25(c). In the present work it may be mentioned that A 2 and S 2 is having mixed mode wear mechanism whereas M 2 reflects ductile nature resulting plastic flow wear mechanism. Sample composition M 2 reflects maximum hardness and have minimum wear rate as compared to all compositions of nanocomposite. Figures 26(a)-(c) represents the worn surface of 3 wt% reinforced nanocomposite. Sample composition A 3 and S 3 shows mixed mode wear mechanism where crack growth, grain pull off and wear debris was observed. Sample composition M 3 reflects brittle nature and hence only large number of sub cracks and micro crack was visible on worn surface.

Conclusion
Out of all fabricated composites sample 2 wt% mixed Al 2 O 3 and Si 3 N 4 exhibited maximum UCS (209.98 MPa) and hardness (56.6 HV) values as compared to the other samples. TEM microstructure for 2 wt% mixed Al 2 O 3 and Si 3 N 4 reinforced sample showed nanometric microstructure with uniform clustering of mixed reinforcement (i.e. Al 2 O 3 and Si 3 N 4 ) while for other samples clustering was not dominant. EBSD results showed the high fraction of HAGBs and low fraction of LAGBs for pure sintered Al sample P and mixed reinforcement composites (M 1 , M 2 and M 3 ) while for rest samples fraction of HAGBs and LAGBs were just vice versa. Texture results for pure sintered Al sample P and mixed reinforcement composites (M 1 , M 2 and M 3 ) showed the strong textures componants near to {112}〈111〉Cu and {110}〈111〉P whereas for other conditions weak textures were observed. Fracture surface morphology of 2 wt% mixed Al 2 O 3 and Si 3 N 4 sample clearly exhibited the mixed modeductile failure mechanism whereas that the surface is entirely covered with dimpled like features with little amount of brittle zone with increasing work hardening ability. Wear rate under under three body wear conditions for 2 wt% mixed Al 2 O 3 and Si 3 N 4 sample composition was found to be least (0.1 mm 3 /min mm 3 /N-cm for 1 kg load and 3.5 mm 3 /N-cm for 2kg load ) out of all other samle compositions. Wear surface morphology of 2 wt% mixed Al 2 O 3 and Si 3 N 4 sample condition revealed the few wear debris and lesser surface intrusions in worn surface causing wear particle plugging and plastic flow wear mechanism.

Acknowledgments
Not applicable.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).