Journal of Materials Research and Technology Journal of Materials Research and Technology
Original Article
Dry sliding wear behaviour of single and dual ceramic reinforcements premixed with Al powder in AA6061 matrix
Ponnarengan Hariharasakthisudhana,, , Swaminathan Joseb, Kondal Manisekarc
a Research Scholar - Sathyabama University, Mechanical Engineering Department, National Engineering College, Kovilpatti, Tamil Nadu, India
b Research Supervisor - Sathyabama University, Mechanical Engineering Department, Loyola-ICAM College of Engineering, Technology, Chennai, Tamil Nadu, India
c Professor - Mechanical Engineering Department, National Engineering College, Kovilpatti, Tamil Nadu, India
Received 21 July 2017, Accepted 08 January 2018

This work highlighted the preparation of a hybrid of AA6061/Al2O3/graphite composite by additionally incorporating Si3N4 nanoparticles in the material system. The reinforcements were premixed with Al metal powder to improve their diffusion into the matrix during manufacturing of composites. The composites were manufactured by stir casting method. The pin on disc wear test was carried out as per ASTM G99 standard. Field emission scanning electron microscope (FESEM) and energy dispersive spectroscopy (EDS) analysis were conducted on worn out surface and subsurface of the composite specimens to determine the wear mechanism. From the results, it was found that the wear behaviour of Al powder premixed reinforcement was superior to unmixed reinforcements. The composites with dual ceramic reinforcement have shown lesser wear rate and coefficient of friction than composites with single ceramic reinforcement. Abrasive wear and grain structure refinement just beneath the worn out surface were noted in composites with Si3N4 nanoparticles. Single ceramic reinforcement of Al premixed Al2O3 was identified with abrasive wear, micro cuts and formation of mechanically mixed layer on the worn out surface. Single ceramic reinforcement of unmixed Al2O3 exhibited abrasive wear and severe plastic deformation on worn out surface. Porosity, particle fracture and particle pull out were noticed in the subsurface of the composites with unmixed single ceramic reinforcement of Al2O3.

Metal–matrix composite, Wear testing, Electron microscopy, Surface analysis
This article is only available in PDF

Inception of using ceramic reinforcement with soft matrix imparted huge development in the field of metal matrix composites (MMC). Tribological behaviour of hard and less dense ceramic particle/fibre dispersed Al based composites were extensively studied by many researchers in micron and nanoscale as Al MMC has wide range of applications. SiC, Al2O3, TiB2, B4C, WC are few synthetic ceramics largely used with Al based matrix. Complex metallic alloys (CMA), industrial waste and agro waste were also experimented with aluminium for wear resistance [1].

AA6061 is currently used in construction, automotive, railroad car bodies, and marine structures. Low wear resistance of AA 6061 should be addressed to increase the applications in structural parts as this property directly affects the structural integrity [2]. Al3O3 is a hard ceramic particle that improved the tribological properties with various Al alloys. Abrasive wear resistance of A2024 alloy was enhanced by increasing the Al2O3 content and size in the composite [3]. Wear resistance of pure Al improved when reinforced with Al2O3 and Al4C3 at increased sliding speeds [4]. Wear mechanisms of alumina reinforced Al alloys with graphite were observed as decohesion, micro cutting and detachment of material in contact surface followed by crack nucleation and propagation in grain boundaries [5]. Delamination was observed with nano-sized Al2O3 reinforced in A2023 matrix [6] and wear rate decreased with increasing content of nano-Al2O3 in A6061 matrix [7]. In Order to further improve the wear resistance of Al alloy/Al2O3/graphite composite, it is required to focus on controlling the particle pullout and successive crack propagation in grain boundary. Particle pullout is attributed to the inefficient bonding in the matrix/reinforcement interface. Metal particles premixed with nano-sized ceramic reinforcement found to provide better dispersion in controlled stirring time with A356 matrix. Improved interfacial bonding was also noted with metal powder mixed reinforcements in matrix [8]. The Si3N4 was found to provide improved fracture toughness with various matrices [9]. The size of the reinforcement particles plays a vital role in deciding the wear and fracture behaviour of the metal matrix composites as it controls the porosity in composites [10].

In this experimental study, it was decided to develop a hybrid of A6061/Al2O3/graphite composite by additionally adding Si3N4 nanoparticles of average size less than 40nm. The reinforcements were premixed with Al powder (99.2% purity). It was intended to analyze the effect of metal mixing in reducing the level of particle pullout and to study the effect of Si3N4 nanoparticles in the A6061/Al2O3/graphite composite to procrastinate the nucleation and propagation of cracks in grain boundaries during wear. Pin on disc wear test was carried out to analyze the tribological behaviour of the composites. The wear behaviour was investigated by varying normal load and sliding speed for constant track length in two categories such as low and high. In case of low category, the normal load was varied as 10N, 20N and 30N for constant sliding speed of 1m/s and track length of 600m. In high category, the sliding speed was varied as 1m/s, 2m/s and 4m/s for constant normal load of 40N and track length of 1000m.

2Experimental procedure2.1Materials preparation

In the context of developing the composites with A6061 matrix, four different mixes of reinforcements were prepared. The Si3N4 nanoparticles used in this research work was commercially purchased. The Si3N4 particles of spherical shape with average size less than 50nm were purchased from Sigma–Aldrich (now Merck) Chemicals. The purity of the powder was ≥98.5%. The morphology of Si3N4 nanoparticle is shown in Fig. 1.

Fig. 1.

Morphology of Si3N4 nanoparticles.

Among all the composites, the amount of reinforcement was 11.5% (120g). In order to explore the potential of metal powder premixing, the compounds Si3N4 and Al2O3, were blended with Al powder and graphite particles. The blend was ball milled for 3h in attrition ball mill with 10min interval per hour. The premixing process was detailed by the authors elsewhere [11]. The composition details of reinforcement added in each composite are tabulated in Table 1.

Table 1.

Composition of reinforcement.

Reinforcement  Composition 
R1  20wt% Al2O3/20wt% Si3N4/55wt% Al/5wt% graphite 
R2  10wt% Al2O3/30wt% Si3N4/55wt% Al/5wt% graphite 
R3  45wt% Al2O3/50wt% Al/5wt% graphite 
R4  95wt% Al2O3/5wt% graphite 

Composites were fabricated by stir casting method. A6061 rods of 1050g were initially melted in the furnace. Molten liquid temperature was maintained at 850°C. Reinforcements were preheated to the temperature of 150°C and they were introduced into the molten metal in three stages where each stage of introduction was followed by stirring at low, middle and top of the melt with graphite stirrer. Two minutes of stirring was done at each depth with the stirring speed of 500rpm. The molten mix was casted as cylindrical rod. The FESEM images of mixed reinforcement 1 are shown in Fig. 2(a), which shows the presence of Al2O3 in the form of flakes and other particles and (b) represents the nanoparticles with small clusters. The EDS report is given in Fig. 3. The XRD result of the prepared composite is shown in Fig. 4, which explains the crystalline nature of the composites. The mechanical properties of composites discussed elsewhere are shown in Table 2.

Fig. 2.

SEM images of (a) blended reinforcement R1 and (b) nanoparticles with small clusters in R1.

Fig. 3.

EDS report of mixed reinforcement R2 [18].

Fig. 4.

XRD result of prepared composite R2.

Table 2.

Mechanical properties of composites.

Composites  Ultimate tensile strength (MPa)  Ultimate compressive strength (MPa)  Vickers hardness (HV)  Ductility (%) 
Cast 1(R1)  131.84  292.6  62.2 
Cast 2 (R2)  153.78  278.49  61.5  8.3 
Cast 3 (R3)  96.16  361.74  66.3  2.5 
Cast 4 (R4)  44.14  298.2  59.9  1.25 
2.2Wear analysis

Pin on disc wear test was carried out as per ASTM G99 standard at room temperature. Cylindrical composite specimens of spherical tip were manufactured for the diameter 6mm and length 45mm. EN31 steel was used as the disc material. Disc was machined for the diameter of 55mm and thickness 10mm. The contact faces of the disc were ground to have average roughness 10μm. Specimens were cleaned with acetone in order to maintain the accuracy of mass measurement before and after the test. Coefficient of friction with respect to time of wear was recorded by the software integrated with testing machine. Wear rate was calculated by measuring mass loss with the help of electronic weighing machine accompanied 1mg accuracy. Test configuration is shown in Fig. 5. Five experiments were conducted in each case and the variability of results was accounted.

Fig. 5.

Test configuration.

Out of the analysis results of wear rate and average coefficient of friction in all the categories of test, one specimen from each cast was selected for further study on subsurface changes in the composite due to wear. Electric discharge machining (EDM) was carried out to cut the specimens longitudinally. Vickers hardness test was conducted to determine the micro hardness variation in the longitudinal direction from the edge of worn out surface in five different spots along the axis. The distance between the indentations was maintained as 0.5mm. Test load was given as 0.5kg and dwell period was set for 10s.

FESEM and EDS analyses were conducted on worn out surfaces to analyze the nature of wear mechanism in each composite. FESEM images were also taken on the longitudinal cut sections to find out the subsurface deformations due to wear.

3Results and discussion3.1Influence of premixed reinforcement in wear rate

Wear rate of the prepared composites in low and high category of wear is represented in Figs. 6 and 7 respectively. As there was no considerable difference in densities of prepared composites, mass loss was used to define the wear rate. Wear rate of composites were increased when normal load was increased as expected. Composites with Al powder premixed reinforcement have shown accountable divergence in wear rate than composites with unmixed reinforcement. Maximum of 15mg/km was observed with premixed reinforcement 3, which was nearly two times lesser than unmixed reinforcement 4 (31.66mg/km) at 30N normal load. The presence of Si3N4 nanoparticle in composite has reduced the wear rate than the reinforcement with the only presence of alumina and graphite. Wear rate was reduced when the volume of Si3N4 nanoparticles increased in the blend of reinforcement. This is attributed to the dispersion and strong interfacial bonding established between fine nano-sized particles and matrix. High chance of fine sized particle pull out from matrix as discussed in Ref. [12] was controlled by the premixing the reinforcement with Al powder. Wear rate of reinforcement 3 was comparatively lower than reinforcement 4. This is characterized by the premixing of Al2O3 and graphite with Al powder which stopped the pull out of Al2O3 particles from the matrix. The increment of wear rate from 20 to 30N normal loads is slightly greater than 10–20N normal load in all the reinforcement. This shows that the increment in normal load increases the shear stress in the contact surfaces which lead to the rise in wear.

Fig. 6.

Representation of wear rate of composites for low category.

Fig. 7.

Representation of wear rate of composites for high category.

In case of high category, wear rate increased as the sliding speed increases. Severe wear was observed with reinforcement 4. This may be due to the rise of friction in tribosystem at high sliding speed which increases the temperature to cause thermal softening. Wear rate increased as thermal softening reduced the shear strength of the material [4]. Weak interfacial bonding may also be another reason as reinforcement 4 was not premixed with Al. At low sliding speeds (i.e. 1m/s and 2m/s), reinforcement 1 and 3 behaved very closely for the wear conditions, whereas, at high sliding speed (i.e. 4m/s) nanoparticle reinforced composite performed better than micron particle composite. Further, reinforcement 2 with large volume of Si3N4 has shown least wear rate in high category as well. This strongly supports the influence of Si3N4 nanoparticle in controlling the particle pull out during wear.

3.2Effect of premixed reinforcement in coefficient of friction

Average COF of the composites at low and high category are shown in Figs. 8 and 9. In case of low category, one can witness the reduction in coefficient of friction as the normal load increases in composites with premixed reinforcements. From the graph, one can notice the low average friction of nanocomposites. This is attributed to the presence of Si3N4 nanoparticle in reinforcement which premixed the hard Al2O3 flakes and reduced the direct contact of composite with counter surface. This effect was further extended in reinforcement 2 with larger amount of Si3N4. The composite with reinforcement 3 showed higher friction than other composites. This is due to Al2O3 flakes strongly bonded with the matrix as they were premixed with Al powders and they cut the counterface which increases the coefficient of friction. It was witnessed that the reinforcement 4 exhibited lower friction than reinforcement 3. This is predicted with the Al2O3 flakes in reinforcement 4 which were unmixed, could not establish the bonding properly with the matrix as a result of that they were pulled out during wear. Even though the reinforcement 4 illustrated lesser average coefficient of friction than reinforcement 3, the wear rate results were opposite between reinforcement 3 and 4. The reason for this conflict is due to the formation of transition layer in between mating surfaces as the strongly bonded alumina in R3 cut the counter face material. The cut out particles forms a mechanically mixed layer (MML) which reduces the wear rate of the material.

Fig. 8.

Representation of average COF of composites for low category.

Fig. 9.

Representation of average COF of composites for high category.

On continuing the behaviour in high category, while increasing the sliding speed from 1 to 2m/s, there is a reduction in average COF. This is attributed to the presence of graphite in all the composites which might have acted as lubricant in between mating surfaces. This phenomenon was encouraged further by thermal softening due to the temperature rise in pin material when the sliding speed was increased. This could be very clearly seen from the result of reinforcement 3. At the sliding speed of 4m/s, the friction was increased. This is because the formed graphite layer was wiped out due to high mechanical action in the interface and the contact was established between the mating materials which enhanced the friction.

3.3Worn surface analysis and mechanism of wear

The FESEM analysis was carried out on worn surfaces and subsurface of four selected composites (one from each cast) which showed exemplary behaviour during wear. The selection was done based on aforementioned results of wear rate and coefficient of friction. The samples were cut according to the dimension required to place them on FESEM equipment. In order to analyze the subsurface of worn out surfaces, the selected samples were cut longitudinally by EDM process. The FESEM images were taken closer to the worn out edge along the subsurface. The EDS analysis was also carried out on the worn out surfaces. The details of selected samples are shown in Table 3.

Table 3.

Details of selected sample for morphology.

Sample  Composite  Condition (normal load/sliding speed/track length)  Wear rate (mg/km)  Average COF 
S1  Cast 1  40N/4m/s/1km  28  0.35 
S2  Cast 2  10N/1m/s/0.6km  1.66  0.40 
S3  Cast 3  40N/2m/s/1km  19  0.46 
S4  Cast 4  40N/4m/s/1km  53  0.53 

It was noted from the micrograph of sample S1 that it has the combined effect of adhesive wear and delamination in few spots. This could be because of small clusters formed in the composite R1, which could nucleate cracks in the grain boundary and led to the delamination on worn out surface as shown in Fig. 10(a). Adhesive wear with small, shallow and parallel grooves along with micro cut was noticed with sample S2 worn out surface [13]. The observed micro cuts were not continuous as reported in Ref. [14]. This is due to the larger volume of Si3N4 nanoparticles which premixed the larger Al2O3 particles to avoid abrasion on counterface. Few wider grooves were also found to show the effect of larger Al2O3 particles as depicted in Fig. 10(b). Deep grooves were observed with sample S3 which explains the abrasive wear. This is because of the only presence of larger premixed Al2O3 particles in matrix [15]. The reason for the abrasion is that the presence of the strongly bonded Al2O3 particles which cut the counterface. This can be witnessed from Fig. 10(c) which shows the protruding particles of Al2O3 in the matrix that was smeared off [3]. It was also proved from the EDS analysis in Fig. 11 that there are traces of Fe and Cr elements on the worn out surface of S3. The presence of O in EDS report represents the possibility for the oxidation of cut out Fe metal particles to form of Fe2O3[16]. This mechanically mixed layer reduces the wear rate as it was formed in between the mating surfaces. The MML was wiped out during high sliding speed and normal load conditions of wear test as discussed by Venkataraman and Sundararajan [17]. Abrasive wear with severe plastic deformation was observed with sample S4 as shown in Fig. 10(d) and (e). This is attributed to the exposure of unmixed Al2O3 particles to the high stress in the interface of matrix and Al2O3 particles. Since the particles are not fused as like Al premixed Al2O3, they tend to nucleate the crack in the interface and subsequently the particles are pulled out.

Fig. 10.

FESEM micrographs of worn surfaces (a) adhesive wear and delamination in S1, (b) grooves with micro cut in S2, (c) protruding Al2O3 particles in S3 (d) and (e) severe plastic deformation in S4 (arrows indicate sliding direction).

Fig. 11.

Results of EDS of composite S3.

3.4Subsurface analysis

The support moulds were prepared to hold the selected composites and the longitudinal cut section was polished with Al2O3 paste on polishing cloths of various grit sizes from coarse to fine. The specimens are shown in Fig. 12. It can be inferred from the results of Vickers micro hardness test represented in Fig. 14, that there was no considerable grain structure changes observed with composite S4 as micro hardness along the axis were almost uniform. The particles pull out closer to the worn out edge of S4 was observed as shown in Fig. 13(d). This must have happened during EDM process. This pull out may be attributed to the nucleation of crack in the grain boundary because of stress transferred during wear. This can be identified from the grain structure of the cavity formed by particle pull out in the Fig. 13(d). This kind of pull out was not observed with other selected composites. Particle fracture (Fig. 13c) and pores (Fig. 13e) were also observed in sample S4, which are predicted with the lack of dispersion of unmixed Al2O3 particles in the matrix. However, Akbari et al. [8] have expressed that there is no consistency in the effect of porosity on wear properties. The composite sample S3 showed high value of hardness at the worn out edge which strongly support the presence of mechanically mixed layer (MML). The samples S1 and S2 have also showed larger values of hardness at the worn out edge. This could be due to the possibility to form graphite layer on the worn out surface which could further delay the fracture of particles [18]. Similar trend of hardness variation along the axis was observed with composite samples S1, S2 and S3 as shown in Fig. 14. It can be predicted based on the coherence of results variation that the considerable plastic deformation and grain structure changes were happened in these three composites for the length of 1mm approximately. As the length extends more than 1mm, the micro hardness reaches the values of as-cast composites. The composites S1 and S2, which contained Si3N4 nanoparticles, were observed with grain structure refinement in nanoscale just beneath the worn out surface as shown in Fig. 13(a) and (b). These changes of grain structure might be due to the heat transferred from the worn out surface during friction coupled with mechanical shear force.

Fig. 12.

Longitudinal cut out specimens for Vickers hardness test.

Fig. 13.

Subsurface FESEM micrographs (a) and (b) grain structure refinement in S1 and S2, (c) particle fracture in S4, (d) particle pull out in S4, and (e) porosity in S4.

Fig. 14.

Representation of Vickers hardness along the subsurface.


Hybrid composite of A6061/Al2O3/graphite with Si3N4 nanoparticles exhibited superior wear properties in both low and high category of wear. The composites with Al premixed reinforcements in A6061 matrix have shown lesser wear rate than composites with unmixed reinforcements. The premixing enhanced the fusion and dispersion of reinforcement particles with matrix. Abrasive wear and the formation of mechanically mixed layer were observed in the composite with the only presence of metal premixed Al2O3 and graphite in matrix. The layer was found to be wiped out at high sliding speed. The composites with reinforcement R1 and R2 have shown adhesive wear along with delamination in few spots. The grain structure refinement in nanoscale was also noted just beneath the worn out surface in the composite with Si3N4 nanoparticles. Abrasive wear and severe plastic deformation was found to be the wear mechanism in the composite without having the metal premixed reinforcement. Particle pullout, porosity, crack nucleation in grain boundary were also noticed with the unmixed reinforcement composite. The micro hardness test expressed that the premixed reinforcement composites have undergone for considerable plastic deformation and grain structure changes in the subsurface to the depth of 1mm approximately due to wear.

Conflicts of interest

The authors declare no conflicts of interest.

M. Oluwatosin,K. Keneth,L. Heath
Aluminium matrix hybrid composites: a review of reinforcement philosophies; mechanical, corrosion and tribological characteristics
Integr Med Res, (2015), pp. 1-12
D. Casellas,A. Beltran,J.M. Prado,A. Larson,A. Romero
Microstructural effects on the dry wear resistance of powder metallurgy Al–Si alloys
Wear, 257 (2004), pp. 730-739
M. Ko
Abrasive wear of Al2O3 particle reinforced 2024 aluminium alloy composites fabricated by vortex method
Composites A, 37 (2006), pp. 457-464
G. Abouelmagd
Hot deformation and wear resistance of P/M aluminium metal matrix composites
J Mater Process Technol, 156 (2004), pp. 1395-1401
A. Baradeswaran,A.E. Perumal
Study on mechanical and wear properties of Al 7075/Al2O3/graphite hybrid composites
Composites B, 56 (2014), pp. 464-471
M. Fellah,M. Abdul,M. Labaiz,O. Assala,A. Iost
Sliding friction and wear performance of the nano-bioceramic α-Al2O3 prepared by high energy milling
Tribiol Int, 91 (2015), pp. 151-159
K. Sahu,R.S. Rana,R. Purohit,D.K. Koli,S.S. Rajpurohit,M. Singh
Wear behavior and micro-structural study of Al/Al2O3 nano-composites before and after heat treatment
Mater Today Proc, 2 (2015), pp. 1892-1900
M.K. Akbari,H.R. Baharvandi,O. Mirzaee
Nano-sized aluminum oxide reinforced commercial casting A356 alloy matrix: evaluation of hardness, wear resistance and compressive strength focusing on particle distribution in aluminum matrix
Composites B, 52 (2013), pp. 262-268
S.R. Wang
Microstructure and fracture characteristic of Mg–Al–Zn–Si3N4 composites
Theor Appl Fract Mech, 46 (2006), pp. 57-69
P.N. Bindumadhavan,H.K. Wah,O. Prabhakar
Dual particle size (DPS) composites: effect on wear and mechanical properties of particulate metal matrix composites
Wear, 248 (2001), pp. 112-120
P. Hariharasakthisudhan,S. Jose
Influence of metal powder premixing on mechanical behavior of dual reinforcement (Al2O3 (mm)/Si3N4 (nm)) in AA6061 matrix
J Alloys Compd, 731 (2018), pp. 100-110
S. Kumar,V. Balasubramanian
Developing a mathematical model to evaluate wear rate of AA7075/SiCp powder metallurgy composites
Wear, 264 (2008), pp. 1026-1034
L.J. Yang
The transient and steady wear coefficients of A6061 aluminium alloy reinforced with alumina particles
Compos Sci Technol, 63 (2003), pp. 575-583
D. Jun,L. Yao-Hui,Y. Si-Rong,L. Wen-Fang
Dry sliding friction and wear properties of Al2O3 and carbon short fibres reinforced Al–12Si alloy hybrid composites
Wear, 257 (2004), pp. 930-940
Z. Zhang,X. Li,E. Almandoz,G. García,H. Dong
Tribology international sliding friction and wear behaviour of titanium-zirconium-molybdenum (TZM) alloy against Al2O3 and Si3N4 balls under several environments and temperatures
Tribiol Int, September (2016), pp. 1
E.M. Sharifi,F. Karimzadeh
Wear behavior of aluminum matrix hybrid nanocomposites fabricated by powder metallurgy
Wear, 271 (2011), pp. 1072-1079
B. Venkataraman,G. Sundararajan
Correlation between the characteristics of the mechanically mixed layer and wear behaviour of aluminium, Al-7075 alloy and Al-MMCs
Wear, 245 (2000), pp. 22-38
J.S.S. Babu,C.G. Kang,H.H. Kim
Dry sliding wear behavior of aluminum based hybrid composites with graphite nanofiber–alumina fiber
Mater Des, 32 (2011), pp. 3920-3925
Corresponding author. (Ponnarengan Hariharasakthisudhan
Copyright © 2018. Brazilian Metallurgical, Materials and Mining Association