- Email: [email protected]
Dry sliding wear behaviour of aluminum syntactic foam
Materials and Design 30 (2009) 2563–2568
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Dry sliding wear behaviour of aluminum syntactic foam D.P. Mondal *, S. Das, Nidhi Jha Advanced Materials and Processes Research Institute (CSIR), Bhopal 462026, India
a r t i c l e
i n f o
Article history: Received 5 August 2008 Accepted 22 September 2008 Available online 6 October 2008 Keywords: Dry sliding wear Aluminum syntactic foam Aluminum composite Porosity
a b s t r a c t Dry sliding wear behaviour of cenosphere reinforced aluminum syntactic foam having density of 1.9 g/cc (around 30% porosity) has been studied using a pin-on-disc apparatus at load range of 1–5 kg and sliding velocity of 2–4 m/s. Coefficient of friction and wear rate of the syntactic foam has been compared with that of 10 wt% SiC reinforced aluminum composite. Scanning electron microscope (SEM) is used to characterize the microstructure and the worn surface for examining the wear mechanism. The craters on the specimen surface due to presence of cenosphere play important role in the wear process. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Aluminum syntactic foam is a new class of materials with important characteristics like light weight, higher specific strength and stiffness, improved high temperature strength, excellent energy absorption characteristics, etc. [1–3]. These materials are formed with the reorganizing hollow-spheres of desired characteristics in desired quantity in the aluminum matrix primarily through melt infiltration technique [4–8]. The melt infiltration technique has the limitation of infiltrating only a thin bed of hollow-spheres and thus, large components could not be made through this process. However, through proper control of the process parameters of stir casting technique, hollow-spheres could be reorganized in the aluminum matrix like that used for dispersion of particulate in liquid melt [9–14]. However, very limited attempts have been made in making aluminum syntactic foam using stir casting technique wherein the reinforcement is cenosphere (a waste product of thermal power plant) [1–3]. The use of cenosphere as reinforcement and the application of stir casting technique for synthesis of syntactic foam would make the materials considerably cheaper than other metallic foams. In addition, these foam materials look to be solid in naked eye even though they contain porosity as high as 30% or more. As these materials exhibit good mechanical properties, they might have reasonably good wear behaviour. A significant amount of work has been carried out on understanding the wear behaviour of dense aluminum matrix composites under dry and lubricating sliding conditions [13–26]. But no attempt has been made so far on examining the sliding wear behaviour of highly porous aluminum syntactic foam. In fact it is suspicious whether these types * Corresponding author. Tel.: +91 755 2417652. E-mail address: [email protected] (D.P. Mondal). 0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2008.09.034
of materials would lead to improved wear resistance due to reduced nominal contact and accumulation of wear debris within the pores and their subsequent compaction, or lead to inferior wear resistance due to higher effective stress on the aluminum matrix and greater possibility of crack nucleation and its propagation [27] due to existence of higher porosity. A few literatures are available on the effect of porosity on the wear behaviour of sintered ceramics and metallic samples [28] where in the amount of porosity is restricted to less than 10%. It is reported that in case of ceramic the wear rate increases with increase in porosity. But in case of metallic sample, the wear rate is hardly affected due to presence of porosity up to 6% [14,29]. Detailed study on the wear behaviour of highly porous material is lacking. In this context, the present paper aims at examining the wear characteristics of highly porous aluminum syntactic foam under dry sliding condition, and finally comparing their wear behaviour with conventional aluminum SiC composites. 2. Experimental Aluminum syntactic foam is prepared by stir casting technique. In this technique aluminum alloy ingot pieces were heated to its molten state. After maintaining the temperature between 750 and 800 °C, a vortex was created using a mechanical stirrer. While stirring was in progress, preheated cenosphere particles were added to the melt. Stirring is continued for about 10 min after addition of cenosphere particles for uniform distribution in the melt. Castings were prepared by pouring the melt into preheated moulds of cylindrical shapes. From these castings samples for wear test of dimensions 8 mm diameter and 28 mm length were prepared. Dry sliding wear tests have been carried out on a pin-on-disc apparatus by sliding a cylindrical pin against the surface of hardened EN24 steel disc under ambient condition. Different loads of
2564
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
1, 3 and 5 kg have been applied on the pin normal to the sliding contact during wear test of the material. The track radius has been kept constant at 65 mm and rotating speed is varied to 294, 441 and 588 rpm corresponding to linear speed of 2, 3 and 4 m/s, respectively. The tests are carried out for a total sliding distance of about 2000 m. The weight was measured on an electronic balance up to 0.001 mg accuracy prior to and after the wear testing. Prior to the wear tests, all the specimens were polished properly. Both the disc and the test specimen have been cleaned well by acetone and then dried under ambient condition prior to and after the tests. The friction force is continuously monitored during the wear test for determining the coefficient of friction. The volumetric wear rate was calculated from the weight loss measurement and expressed in terms of m3/m. 2.1. Microstructural examination For microstructural examination samples are mechanically polished using standard metallographic technique and then etched with Keller’s reagent. The etched samples were examined in scanning electron microscope (SEM). Wear surfaces were also examined in SEM. Prior to SEM examinations, samples were sputtered with gold. 3. Results and discussion 3.1. Material and microstructure The microstructure of aluminum syntactic foam is shown in Fig. 1a. It shows that cenosphere particles are uniformly distrib-
uted within the metal matrix. The average particle size of cenosphere in syntactic foam is noted to be 85 ± 5 lm. Higher magnification micrograph of syntactic foam shows reasonably good bonding between cenosphere and the matrix Fig. 1b. The shell wall of the cenosphere was noted to be around 5–8 lm. The microstructure of SiC reinforced Al-composites is shown in Fig. 1c. It depicts that the SiC particles are angular in nature and distributed uniformly in the matrix. 3.2. Coefficient of friction The variation of coefficient of friction as a function of sliding distance at different applied load is shown in Fig. 2. It is observed that coefficient of friction varies in oscillating fashion within a certain range. In case of 3 and 5 kg load, the range of oscillation is very low and the coefficient of friction varies in the range of 0.06–0.08 and 0.05–0.07, respectively. On the other hand, at lower applied load (1 kg), the coefficient of friction is noted to be considerably higher and it also oscillates at wider range of variation, i.e., 0.09– 0.14. It may further be noted that the maximum coefficient of friction is attained at 1 kg applied load. Similar kind of trend of variation in coefficient of friction with sliding distance was noted at other sliding speeds and applied load. But the magnitudes of coefficient of friction are noted to be varying with applied loads and sliding velocities. The average values of coefficient of friction are recorded from the tests and are plotted as a function of sliding velocity (Fig. 3) and applied load (Fig. 4). It is evident from Fig. 3, that coefficient of friction decreases with increasing applied load especially for sliding velocity of 3 and 4 m/s. In case of sliding velocity of 2 m/s, the coefficient of friction reduced to the mini-
Fig. 1. Microstructure of aluminum syntactic foam: (a) distribution of cenospheres in the matrix, (b) interface between cenosphere and the matrix and (c) microstructure of Al–SiC composite.
2565
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
with increase in sliding velocity and reached to the maximum value at sliding velocity of 3 m/s. In case of applied load of 1 kg, the coefficient of friction reduced significantly after attaining the maximum when the velocity increased further to 4 m/s. But in case of 3 kg applied load, the coefficient of friction remains unchanged after attaining the maxima when the sliding velocity increases to 4 m/s. In case of 5 kg applied load, the coefficient of friction remains almost invariant with sliding velocity.
L=1kg L=3kg L=5kg
0.12
0.08
3.3. Wear rate
0.04
0 0
500
1000
1500
2000
2500
Sliding distance (m) Fig. 2. Variation of coefficient of friction as a function of sliding distance at different applied loads and sliding velocity of 3 m/s.
0.14
v=3m/s
0.1
v=4m/s
0.08
12
L=1kg
0.06 10 0.04
3
Coefficient of friction
v=2m/s 0.12
The wear rate as a function of velocity is shown in Fig. 5. It is observed from this figure that the wear rate decreases with increasing sliding velocity. The maximum wear rate of 11.06 10 12 m3/m occurs at an applied load of 5 kg and sliding velocity of 2 m/s. With further increase in sliding velocity, the wear rate decreases monotonously. But in case of applied load of 1 and 3 kg the reduction in wear rate with increasing sliding velocity is almost similar with a minimum value of 0.8 10 12 m3/m at sliding velocity of 4 m/s. It is observed from Fig. 6 that the wear rate remains almost unchanged up to applied load of 3 kg and then increases sharply when load increased from 3 to 5 kg. There is significant increase in wear rate i.e. from 1.38 10 12 to 11.1 10 12 m3/m with increase in applied load from 3 to 5 kg at a sliding velocity of
Wear rate (10-12m /m)
Coefficient of friction
0.16
0.02 0
0
2
4
6
Load(kg) Fig. 3. Variation of average coefficient of friction as a function of applied load at different sliding velocities.
L=3kg L=5kg
8 6 4 2 0 0
1
0.14 L=1kg 0.12
4
5
Fig. 5. Variation of wear rate as a function of sliding velocity at different applied loads.
L=3kg L=5kg
0.1 12
0.08
v=2m/s
0.06 0.04 0.02 0 0
1
2
3
4
5
velocity (m/s) Fig. 4. Variation of average coefficient of friction as a function of sliding velocity at different applied loads.
mum of 0.04 at an applied load of 3 kg and with further increase in applied load, it increases to 0.06. It is evident from Fig. 4 that, in case of 1 and 3 kg applied load, the coefficient of friction increases
Wear rate (10-12m3/m)
Coefficient of friction
3
2
velocity m/s
v=3m/s
10
v=4m/s 8 6 4 2 0 0
1
2
3
4
5
6
Load (kg) Fig. 6. Variation of wear rate as a function of applied load at different sliding velocities.
2566
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
Table 1 Comparison of wear rate and coefficient of friction in Al–SiC composites with that of aluminum syntactic foam. 12
m3/m)
Load (kg)
Speed (m/s)
Coefficient of friction
Wear rate (10
SF
Composite
SF
Composite
1 3 5
3 3 3
0.12 0.075 0.04
0.6 0.6 0.6
0.5 0.5 9.5
0.4 0.4 3.0
2 m/s. This indicates transition in wear mechanism when the applied load changes from 3 to 5 kg which could be understand from wear surface examination.
This could be due to several reasons. Because of porous structure, the pores acts as craters which lead to reduced nominal contact. These craters help in accumulating the wear debris which is subsequently due to higher contact load and temperature gets compacted with in the craters. Thus, the fraction of material removed from the surface gets reduced. The cenosphere, contains shells of alumino silicates which are fragile and gets fragmented into very fine particles and mixed over the wear surface because of combined effect of frictional heating and surface plastic deformation and material flow, and thus, make a stronger mechanically mixed layer (MML) especially at higher applied load and sliding speed. At lower applied load, wear debris might be fewer in number and frictional heating is low. As a result, large amount of wear
Fig. 7. Worn surface of aluminum syntactic foam: (a) at a load of 3 kg and sliding velocity of 2 m/s, (b) magnified microstructure of worn surfaces at 3 kg load and 2 m/s sliding velocity, (c) at a load of 3 kg and sliding velocity of 3 m/s, (d) at a load of 3 kg and sliding velocity of 4 m/s, (e) at a load of 5 kg and sliding velocity of 4 m/s and (f) at a load of 5 kg and sliding velocity of 4 m/s showing delamination of compacted wear debris from the craters (cenospheres).
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
debris gets removed from the wear surface and only a minor fraction gets accumulated and compacted at the craters or cenosphere sites. Thus at lower applied load surface roughness is more and leads to higher coefficient of friction. Additionally surface materials are harder due to lower frictional heating and thus greater frictional force is required for sliding. At higher applied load, frictional heating is considerably higher and relatively more number of wear debris having larger size were generated which are accommodated or entrapped and get compacted at the craters of cenospheres. This leads to greater extent of plastic flow and relatively smoother surface, and which finally resulting in reduction in coefficient of friction. On the contrary, at higher applied load there is also greater tendency of delamination of wear debris especially from the craters because of weak bonding between the debris compacted in the crater and the crater surface, large extent of surface crack generation and localized fusion (because of higher frictional heating) which leads to higher degree of adhesive wear in addition to abrasive kind of wear. As a result the wear rate increases significantly when the applied load is increased from 3 to 5 kg. When the applied load is limited to 3 kg, the wear rate is primarily governed by abrasive type of wear and a large extent of wear debris gets accumulated at the craters. This type of phenomenon could be understood from wear surface examination. 3.4. Comparison with Al–SiC composite The wear rate and average coefficient of friction of Al syntactic foam and 10 wt% Al–SiC composite have been compared in Table 1. It is evident from this table that the wear rate highly porous Al syntactic is comparable to that of Al–SiC composite especially at applied load up to 3 kg. At higher applied load (5 kg), the former one suffers from considerably higher wear rate than that of the later one. However, both the material shows the trend of sudden increase in wear rate due to increase in load from 3 to 5 kg. This signifies that transition in wear mechanisms take place when applied load increased above 3 kg. As both the composites have the same matrix material, it is expected that matrix plays an important role on transition of wear mechanism. On the contrary to the wear rate, coefficient of friction of the SiC reinforced composites always one order greater than that of cenosphere reinforced aluminum syntactic foam. This may be attributed to the fact that hard, rigid, angular and protruded SiC particles might cause greater abrasive action to the counter surface and these SiC particles also provides greater resistance for slipping action as they can maintain greater strength even at higher temperature. On the other hand, the cenosphere are hollow with shell thickness of 5–8 lm gets fragmented and does not cause hardly any abrasive action to the counter surface. Additionally, these particles get easily mixed into the wear surface material during wear to make MML. The slipping action due to easy surface material flow on account of frictional heating and sliding is expected to be more in syntactic foam. All these factors lead to considerably less coefficient of friction in aluminum syntactic foam as compared to that of Al–SiC composite. 3.5. Wear surface The worn surface of syntactic foam when tested at a load of 3 kg and sliding velocity of 2 m/s is shown in Fig. 7a. It depicts continuous wear grooves and craters (marked ‘c’). The craters are the sites of cenosphere present in the syntactic foam. The presence of wear grooves demonstrates abrasive type of wear mechanism prevailing under such conditions. Higher magnification photograph of these worn surfaces is shown in Fig. 7b. It also depicts flow of material on the craters (cenosphere) and accumulation of wear debris within the craters (marked ‘A’). It demonstrates that sliding wear to some extent is also prevailing under such wear
2567
conditions. Because of combined action of temperature rise, flow of surface material, accumulation of wear debris in the craters and applied load, mechanically mixed layers on the wear surface is generated which contains fine cenosphere shell (white in colour), matrix materials, counter surface material and oxides of the aluminum [30]. The cenosphere shells are primarily made of alumina silicates and during wear process those cells get fragmented and mixed with the matrix on the wear surface and leads to increase in surface hardness. The same sample when tested at a sliding velocity of 3 m/s, the wear surface is characterized with continuous and deeper wear grooves along with craters (marked ‘c’), which are relatively more, elongated (Fig. 7c). This figure also shows greater extent of wear debris accumulation in the craters (marked ‘A’), which make the surface relatively smoother. The surface also demonstrates formation of MML to a greater extent. At higher sliding velocity (4 m/s), the wear surface Fig. 7d is observed to be almost similar to that observed at a sliding velocity of 3 m/s. Because of relatively smoother surface and softening of surface material the coefficient of friction decreases with sliding velocity and applied load. However when the applied load is increased to 5 kg, the worn surface is characterized in terms of the large extent of surface cracks (arrow marked), greater extent of material flow and smearing tendency Fig. 7e. The craters are almost filled with the wear debris and the surface becomes smoother resulting in lower coefficient of friction. At the same time large extent of materials gets removed due to greater degree of delaminating wear in addition to abrasive wear. Fig. 7f shows the detachment of compacted wear debris from the cenosphere craters due to combined action of localized fusion, higher extent of crack generation and weak bonding between cenosphere shells and the compacted wear debris, which leads to fresh cenosphere craters (marked ‘c’). This figure also clearly demonstrates entrapment and accumulation of wear debris with in the cenosphere craters.
4. Conclusions Cenosphere reinforced aluminum syntactic foam exhibits reasonably good wear behaviour even though these are highly porous in nature. The micropores (cenospheres) acts as craters which reduced the nominal contact with the counter surface and helps in accumulation in wear debris generated during wear process. At lower applied load and slower sliding speed frictional heating is low and finer wear debris are generated and as a result the possibility of accumulation and compaction of wear debris at the craters is less and thus the surface becomes relatively rough which results in higher coefficient of friction. At higher applied load, coarser wear debris generated during wear has the greater tendency to be accumulated and compacted at these craters because of combined action of frictional heating, greater degree of material flow and applied load. These would lead to smoother surface and greater degree of sliding action which leads to reduction in coefficient of friction. At higher applied load, irrespective of sliding speed, greater degree of adhesion take place and this leads to delaminating wear especially from the craters because of greater extent of crack generation and weak bonding between compacted wear debris and crater surface. This leads to sharp transition in wear rate due to increase in applied load from 3 to 5 kg. As compared to Al–SiC composites, aluminum syntactic foam exhibits significantly less coefficient of friction which may be attributed to greater abrasive action of protruded SiC particles over the counter surface. The wear rate at lower applied load is noted to be almost same in both the materials. At higher applied load, aluminum syntactic foam exhibits greater wear rate as compared to Al–SiC composite because of the higher strength and less extent of adhesive wear due to presence of rigid and hard SiC particles.
2568
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
References [1] Kiser M, He YM, Zok FW. The mechanical response of ceramic microballon reinforced aluminium matrix composite under compression loading. Acta Mater 1999;47:2685. [2] Kim HS, Plubrai P. Manufacturing and failure mechanism of syntactic foam under compression. Compos Pt A 2004;35:1009. [3] Balch DK, O’Dwyer JG, Davis GR, Cady CM, Gray GT, Dunand DC. Plasticity and damage in aluminium syntactic foams deformed under dynamic and quasi static condition. Mater Sci Eng A 2005;391:408. [4] Palmer RA, Gao K, Doam TM, Green L, Cavallaro G. Pressure infiltrated syntactic foams – process development and mechanical properties. Mater Sci Eng 2007;464:85–92. [5] Wu CML, Han GW. Synthesis of an Al2O3/Al co-continuous composite by reactive melt infiltration. Mater Charact 2007;58:416–22. [6] Mizumoto Masayuki, Ohga Takeshi, Kagawa Akio. Characterization of fiberreinforced metal matrix composites fabricated by low-pressure infiltration process. Mater Sci Eng 2005;413–414:521–6. [7] Rao B Srinivasa, Jayaram V. Pressure less infiltration of Al–Mg based alloys into preforms: mechanisms and phenomenology. Acta Mater Al2O3 2001;49:2373–85. [8] Cook AJ, Werner PS. Pressure infiltration casting of metal matrix composites. Mater Sci Eng A 1991;144:189–206. [9] Ceschini L, Bosi C, Casagrande A, Garagnani GL. Effect of thermal treatment and recycling on the tribological behaviour of an AlSiMg–SiCp composite. Wear 2001;251:1377–85. [10] Singh M, Prasad BK, Mondal DP, Jha AK. Dry sliding wear behaviour of an aluminium alloy–granite particle composite. Tribol Int 2001;34:557–67. [11] Modi OP, Rathod S, Prasad BK, Jha AK, Dixit G. The influence of alumina particle dispersion and test parameter on dry sliding wear behaviour of zinc-based alloy. Tribol Int 2007;40:1137–46. [12] Uyyuru RK, Surappa MK, Brusethaug S. Effect of reinforcement volume fraction and size distribution on the tribological behaviour of Al composite/brake pad tribo-couple. Wear 2006;260:1248–55. [13] Sudrashan, Surappa MK. Dry sliding wear of fly ash particle reinforced A356 Al-composites. Wear 2008;265:349–60. [14] Hamid A Abdulhaqq, Gosh PK, Jain SC, Ray Subrata. Influence of porosity and particles content on dry sliding wear of cast in situ Al(Ti)–Al2O3(TiO2) composite. Wear 2008;265:349–60. [15] Ranganath G, Sharma SC, Krishna M. Dry sliding wear of garnet reinforced zinc/aluminium metal matrix composites. Wear 2001;251:1408–13.
[16] Wang Y, Rainforth WM, Jones H, Lieblich M. Dry wear behaviour and its relation to microstructure of novel 6092 aluminium alloy–Ni3Al powder metallurgy composite. Wear 2001;251:1421–32. [17] Hui Di-Zhou, Chen Jian-Min, Jia Jun-Hong. Investigation of the tribological behaviour of an aluminium alloy with embedded materials. Mater Sci Eng A 2001;302:222–6. [18] Bermudez MD, Nicolas G Martinez, Carrion FJ, Mateo Martinez, Rodriguez JA, et al. Dry and lubricated wear resistance of mechanically alloyed aluminium base sintered composite. Wear 2001;248:178–86. [19] Rodriguez J, Poza P, Garrido MA, Rico A. Dry sliding wear behaviour of aluminium–lithium alloys reinforced with SiC particles. Wear 2007;262: 292–300. [20] Zhang Shaoyang, Wang Fuping. Comparison of friction and wear performance of brake material dry sliding against two aluminium matrix composites reinforced with different SiC particles. J Mater Proc Technol 2007;182:122–7. [21] Ahlatci Hayrettin, Kocer Tolga, Candan Ercan, Cimenoglu Huseyin. Wear behaviour of Al/(Al2O3p + SiCp) hybrid composites. Tribol Int 2006;39:213–20. [22] Cao GH, Wu SQ, Liu JM, Liu ZG. Wear resistance mechanism of an Al–12Si alloy reinforced with aluminisilicate short fibres. Tribol Int 1999;32:721–4. [23] Tjong SC, Wu SQ, Liao HC. Wear behaviour of an Al–12% Si alloy reinforced with low volume fraction of SiC particles. Compos Sci Tech 1998;57:1551–8. [24] Sahin Y, Murphy ST. He effect of sliding speed and microstructure on the dry wear properties of metal matrix composites. Wear 1998;214:98–106. [25] Sun Zhiqiang, Zhang Di, Li Guobin. Effect of dry sliding wear behaviour of silicon particles reinforced aluminium matrix composite. Mater Des 2005;26:454–8. [26] Alpas AT, Embury JD. Sliding and abrasive wear resistance of an aluminium– SiC particle reinforced composite. Scripta Metall Mater 1990:931–5. [27] Ray S, Fishman SG, Dhingra AK, editors. Symposium on advances in cast metal matrix composite: challenges in processing and design. ASM International; 1988. p. 77. [28] Vardavoulias M, Jouannt Tresy C, Jeandin M. Sliding wear behaviour of ceramic particle reinforced high-speed steel obtained by powder metallurgy. Wear 1993;165:141–9. [29] Wang Junan, He Yeng, Danninger Herbert. Influence of porosity on the sliding wear behaviour of sintered Fe–1.5Mo–0.7C steels. J Mater Eng Perf 2003;12:339–44. [30] Mondal DP, Das S, Rao RN, Singh M. Effect of SiC addition and running-in-wear on sliding wear behaviour of Al–Zn–Mg aluminium alloy. Mater Sci Eng A 2005;402:307–19.
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Dry sliding wear behaviour of aluminum syntactic foam D.P. Mondal *, S. Das, Nidhi Jha Advanced Materials and Processes Research Institute (CSIR), Bhopal 462026, India
a r t i c l e
i n f o
Article history: Received 5 August 2008 Accepted 22 September 2008 Available online 6 October 2008 Keywords: Dry sliding wear Aluminum syntactic foam Aluminum composite Porosity
a b s t r a c t Dry sliding wear behaviour of cenosphere reinforced aluminum syntactic foam having density of 1.9 g/cc (around 30% porosity) has been studied using a pin-on-disc apparatus at load range of 1–5 kg and sliding velocity of 2–4 m/s. Coefficient of friction and wear rate of the syntactic foam has been compared with that of 10 wt% SiC reinforced aluminum composite. Scanning electron microscope (SEM) is used to characterize the microstructure and the worn surface for examining the wear mechanism. The craters on the specimen surface due to presence of cenosphere play important role in the wear process. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Aluminum syntactic foam is a new class of materials with important characteristics like light weight, higher specific strength and stiffness, improved high temperature strength, excellent energy absorption characteristics, etc. [1–3]. These materials are formed with the reorganizing hollow-spheres of desired characteristics in desired quantity in the aluminum matrix primarily through melt infiltration technique [4–8]. The melt infiltration technique has the limitation of infiltrating only a thin bed of hollow-spheres and thus, large components could not be made through this process. However, through proper control of the process parameters of stir casting technique, hollow-spheres could be reorganized in the aluminum matrix like that used for dispersion of particulate in liquid melt [9–14]. However, very limited attempts have been made in making aluminum syntactic foam using stir casting technique wherein the reinforcement is cenosphere (a waste product of thermal power plant) [1–3]. The use of cenosphere as reinforcement and the application of stir casting technique for synthesis of syntactic foam would make the materials considerably cheaper than other metallic foams. In addition, these foam materials look to be solid in naked eye even though they contain porosity as high as 30% or more. As these materials exhibit good mechanical properties, they might have reasonably good wear behaviour. A significant amount of work has been carried out on understanding the wear behaviour of dense aluminum matrix composites under dry and lubricating sliding conditions [13–26]. But no attempt has been made so far on examining the sliding wear behaviour of highly porous aluminum syntactic foam. In fact it is suspicious whether these types * Corresponding author. Tel.: +91 755 2417652. E-mail address: [email protected] (D.P. Mondal). 0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2008.09.034
of materials would lead to improved wear resistance due to reduced nominal contact and accumulation of wear debris within the pores and their subsequent compaction, or lead to inferior wear resistance due to higher effective stress on the aluminum matrix and greater possibility of crack nucleation and its propagation [27] due to existence of higher porosity. A few literatures are available on the effect of porosity on the wear behaviour of sintered ceramics and metallic samples [28] where in the amount of porosity is restricted to less than 10%. It is reported that in case of ceramic the wear rate increases with increase in porosity. But in case of metallic sample, the wear rate is hardly affected due to presence of porosity up to 6% [14,29]. Detailed study on the wear behaviour of highly porous material is lacking. In this context, the present paper aims at examining the wear characteristics of highly porous aluminum syntactic foam under dry sliding condition, and finally comparing their wear behaviour with conventional aluminum SiC composites. 2. Experimental Aluminum syntactic foam is prepared by stir casting technique. In this technique aluminum alloy ingot pieces were heated to its molten state. After maintaining the temperature between 750 and 800 °C, a vortex was created using a mechanical stirrer. While stirring was in progress, preheated cenosphere particles were added to the melt. Stirring is continued for about 10 min after addition of cenosphere particles for uniform distribution in the melt. Castings were prepared by pouring the melt into preheated moulds of cylindrical shapes. From these castings samples for wear test of dimensions 8 mm diameter and 28 mm length were prepared. Dry sliding wear tests have been carried out on a pin-on-disc apparatus by sliding a cylindrical pin against the surface of hardened EN24 steel disc under ambient condition. Different loads of
2564
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
1, 3 and 5 kg have been applied on the pin normal to the sliding contact during wear test of the material. The track radius has been kept constant at 65 mm and rotating speed is varied to 294, 441 and 588 rpm corresponding to linear speed of 2, 3 and 4 m/s, respectively. The tests are carried out for a total sliding distance of about 2000 m. The weight was measured on an electronic balance up to 0.001 mg accuracy prior to and after the wear testing. Prior to the wear tests, all the specimens were polished properly. Both the disc and the test specimen have been cleaned well by acetone and then dried under ambient condition prior to and after the tests. The friction force is continuously monitored during the wear test for determining the coefficient of friction. The volumetric wear rate was calculated from the weight loss measurement and expressed in terms of m3/m. 2.1. Microstructural examination For microstructural examination samples are mechanically polished using standard metallographic technique and then etched with Keller’s reagent. The etched samples were examined in scanning electron microscope (SEM). Wear surfaces were also examined in SEM. Prior to SEM examinations, samples were sputtered with gold. 3. Results and discussion 3.1. Material and microstructure The microstructure of aluminum syntactic foam is shown in Fig. 1a. It shows that cenosphere particles are uniformly distrib-
uted within the metal matrix. The average particle size of cenosphere in syntactic foam is noted to be 85 ± 5 lm. Higher magnification micrograph of syntactic foam shows reasonably good bonding between cenosphere and the matrix Fig. 1b. The shell wall of the cenosphere was noted to be around 5–8 lm. The microstructure of SiC reinforced Al-composites is shown in Fig. 1c. It depicts that the SiC particles are angular in nature and distributed uniformly in the matrix. 3.2. Coefficient of friction The variation of coefficient of friction as a function of sliding distance at different applied load is shown in Fig. 2. It is observed that coefficient of friction varies in oscillating fashion within a certain range. In case of 3 and 5 kg load, the range of oscillation is very low and the coefficient of friction varies in the range of 0.06–0.08 and 0.05–0.07, respectively. On the other hand, at lower applied load (1 kg), the coefficient of friction is noted to be considerably higher and it also oscillates at wider range of variation, i.e., 0.09– 0.14. It may further be noted that the maximum coefficient of friction is attained at 1 kg applied load. Similar kind of trend of variation in coefficient of friction with sliding distance was noted at other sliding speeds and applied load. But the magnitudes of coefficient of friction are noted to be varying with applied loads and sliding velocities. The average values of coefficient of friction are recorded from the tests and are plotted as a function of sliding velocity (Fig. 3) and applied load (Fig. 4). It is evident from Fig. 3, that coefficient of friction decreases with increasing applied load especially for sliding velocity of 3 and 4 m/s. In case of sliding velocity of 2 m/s, the coefficient of friction reduced to the mini-
Fig. 1. Microstructure of aluminum syntactic foam: (a) distribution of cenospheres in the matrix, (b) interface between cenosphere and the matrix and (c) microstructure of Al–SiC composite.
2565
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
with increase in sliding velocity and reached to the maximum value at sliding velocity of 3 m/s. In case of applied load of 1 kg, the coefficient of friction reduced significantly after attaining the maximum when the velocity increased further to 4 m/s. But in case of 3 kg applied load, the coefficient of friction remains unchanged after attaining the maxima when the sliding velocity increases to 4 m/s. In case of 5 kg applied load, the coefficient of friction remains almost invariant with sliding velocity.
L=1kg L=3kg L=5kg
0.12
0.08
3.3. Wear rate
0.04
0 0
500
1000
1500
2000
2500
Sliding distance (m) Fig. 2. Variation of coefficient of friction as a function of sliding distance at different applied loads and sliding velocity of 3 m/s.
0.14
v=3m/s
0.1
v=4m/s
0.08
12
L=1kg
0.06 10 0.04
3
Coefficient of friction
v=2m/s 0.12
The wear rate as a function of velocity is shown in Fig. 5. It is observed from this figure that the wear rate decreases with increasing sliding velocity. The maximum wear rate of 11.06 10 12 m3/m occurs at an applied load of 5 kg and sliding velocity of 2 m/s. With further increase in sliding velocity, the wear rate decreases monotonously. But in case of applied load of 1 and 3 kg the reduction in wear rate with increasing sliding velocity is almost similar with a minimum value of 0.8 10 12 m3/m at sliding velocity of 4 m/s. It is observed from Fig. 6 that the wear rate remains almost unchanged up to applied load of 3 kg and then increases sharply when load increased from 3 to 5 kg. There is significant increase in wear rate i.e. from 1.38 10 12 to 11.1 10 12 m3/m with increase in applied load from 3 to 5 kg at a sliding velocity of
Wear rate (10-12m /m)
Coefficient of friction
0.16
0.02 0
0
2
4
6
Load(kg) Fig. 3. Variation of average coefficient of friction as a function of applied load at different sliding velocities.
L=3kg L=5kg
8 6 4 2 0 0
1
0.14 L=1kg 0.12
4
5
Fig. 5. Variation of wear rate as a function of sliding velocity at different applied loads.
L=3kg L=5kg
0.1 12
0.08
v=2m/s
0.06 0.04 0.02 0 0
1
2
3
4
5
velocity (m/s) Fig. 4. Variation of average coefficient of friction as a function of sliding velocity at different applied loads.
mum of 0.04 at an applied load of 3 kg and with further increase in applied load, it increases to 0.06. It is evident from Fig. 4 that, in case of 1 and 3 kg applied load, the coefficient of friction increases
Wear rate (10-12m3/m)
Coefficient of friction
3
2
velocity m/s
v=3m/s
10
v=4m/s 8 6 4 2 0 0
1
2
3
4
5
6
Load (kg) Fig. 6. Variation of wear rate as a function of applied load at different sliding velocities.
2566
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
Table 1 Comparison of wear rate and coefficient of friction in Al–SiC composites with that of aluminum syntactic foam. 12
m3/m)
Load (kg)
Speed (m/s)
Coefficient of friction
Wear rate (10
SF
Composite
SF
Composite
1 3 5
3 3 3
0.12 0.075 0.04
0.6 0.6 0.6
0.5 0.5 9.5
0.4 0.4 3.0
2 m/s. This indicates transition in wear mechanism when the applied load changes from 3 to 5 kg which could be understand from wear surface examination.
This could be due to several reasons. Because of porous structure, the pores acts as craters which lead to reduced nominal contact. These craters help in accumulating the wear debris which is subsequently due to higher contact load and temperature gets compacted with in the craters. Thus, the fraction of material removed from the surface gets reduced. The cenosphere, contains shells of alumino silicates which are fragile and gets fragmented into very fine particles and mixed over the wear surface because of combined effect of frictional heating and surface plastic deformation and material flow, and thus, make a stronger mechanically mixed layer (MML) especially at higher applied load and sliding speed. At lower applied load, wear debris might be fewer in number and frictional heating is low. As a result, large amount of wear
Fig. 7. Worn surface of aluminum syntactic foam: (a) at a load of 3 kg and sliding velocity of 2 m/s, (b) magnified microstructure of worn surfaces at 3 kg load and 2 m/s sliding velocity, (c) at a load of 3 kg and sliding velocity of 3 m/s, (d) at a load of 3 kg and sliding velocity of 4 m/s, (e) at a load of 5 kg and sliding velocity of 4 m/s and (f) at a load of 5 kg and sliding velocity of 4 m/s showing delamination of compacted wear debris from the craters (cenospheres).
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
debris gets removed from the wear surface and only a minor fraction gets accumulated and compacted at the craters or cenosphere sites. Thus at lower applied load surface roughness is more and leads to higher coefficient of friction. Additionally surface materials are harder due to lower frictional heating and thus greater frictional force is required for sliding. At higher applied load, frictional heating is considerably higher and relatively more number of wear debris having larger size were generated which are accommodated or entrapped and get compacted at the craters of cenospheres. This leads to greater extent of plastic flow and relatively smoother surface, and which finally resulting in reduction in coefficient of friction. On the contrary, at higher applied load there is also greater tendency of delamination of wear debris especially from the craters because of weak bonding between the debris compacted in the crater and the crater surface, large extent of surface crack generation and localized fusion (because of higher frictional heating) which leads to higher degree of adhesive wear in addition to abrasive kind of wear. As a result the wear rate increases significantly when the applied load is increased from 3 to 5 kg. When the applied load is limited to 3 kg, the wear rate is primarily governed by abrasive type of wear and a large extent of wear debris gets accumulated at the craters. This type of phenomenon could be understood from wear surface examination. 3.4. Comparison with Al–SiC composite The wear rate and average coefficient of friction of Al syntactic foam and 10 wt% Al–SiC composite have been compared in Table 1. It is evident from this table that the wear rate highly porous Al syntactic is comparable to that of Al–SiC composite especially at applied load up to 3 kg. At higher applied load (5 kg), the former one suffers from considerably higher wear rate than that of the later one. However, both the material shows the trend of sudden increase in wear rate due to increase in load from 3 to 5 kg. This signifies that transition in wear mechanisms take place when applied load increased above 3 kg. As both the composites have the same matrix material, it is expected that matrix plays an important role on transition of wear mechanism. On the contrary to the wear rate, coefficient of friction of the SiC reinforced composites always one order greater than that of cenosphere reinforced aluminum syntactic foam. This may be attributed to the fact that hard, rigid, angular and protruded SiC particles might cause greater abrasive action to the counter surface and these SiC particles also provides greater resistance for slipping action as they can maintain greater strength even at higher temperature. On the other hand, the cenosphere are hollow with shell thickness of 5–8 lm gets fragmented and does not cause hardly any abrasive action to the counter surface. Additionally, these particles get easily mixed into the wear surface material during wear to make MML. The slipping action due to easy surface material flow on account of frictional heating and sliding is expected to be more in syntactic foam. All these factors lead to considerably less coefficient of friction in aluminum syntactic foam as compared to that of Al–SiC composite. 3.5. Wear surface The worn surface of syntactic foam when tested at a load of 3 kg and sliding velocity of 2 m/s is shown in Fig. 7a. It depicts continuous wear grooves and craters (marked ‘c’). The craters are the sites of cenosphere present in the syntactic foam. The presence of wear grooves demonstrates abrasive type of wear mechanism prevailing under such conditions. Higher magnification photograph of these worn surfaces is shown in Fig. 7b. It also depicts flow of material on the craters (cenosphere) and accumulation of wear debris within the craters (marked ‘A’). It demonstrates that sliding wear to some extent is also prevailing under such wear
2567
conditions. Because of combined action of temperature rise, flow of surface material, accumulation of wear debris in the craters and applied load, mechanically mixed layers on the wear surface is generated which contains fine cenosphere shell (white in colour), matrix materials, counter surface material and oxides of the aluminum [30]. The cenosphere shells are primarily made of alumina silicates and during wear process those cells get fragmented and mixed with the matrix on the wear surface and leads to increase in surface hardness. The same sample when tested at a sliding velocity of 3 m/s, the wear surface is characterized with continuous and deeper wear grooves along with craters (marked ‘c’), which are relatively more, elongated (Fig. 7c). This figure also shows greater extent of wear debris accumulation in the craters (marked ‘A’), which make the surface relatively smoother. The surface also demonstrates formation of MML to a greater extent. At higher sliding velocity (4 m/s), the wear surface Fig. 7d is observed to be almost similar to that observed at a sliding velocity of 3 m/s. Because of relatively smoother surface and softening of surface material the coefficient of friction decreases with sliding velocity and applied load. However when the applied load is increased to 5 kg, the worn surface is characterized in terms of the large extent of surface cracks (arrow marked), greater extent of material flow and smearing tendency Fig. 7e. The craters are almost filled with the wear debris and the surface becomes smoother resulting in lower coefficient of friction. At the same time large extent of materials gets removed due to greater degree of delaminating wear in addition to abrasive wear. Fig. 7f shows the detachment of compacted wear debris from the cenosphere craters due to combined action of localized fusion, higher extent of crack generation and weak bonding between cenosphere shells and the compacted wear debris, which leads to fresh cenosphere craters (marked ‘c’). This figure also clearly demonstrates entrapment and accumulation of wear debris with in the cenosphere craters.
4. Conclusions Cenosphere reinforced aluminum syntactic foam exhibits reasonably good wear behaviour even though these are highly porous in nature. The micropores (cenospheres) acts as craters which reduced the nominal contact with the counter surface and helps in accumulation in wear debris generated during wear process. At lower applied load and slower sliding speed frictional heating is low and finer wear debris are generated and as a result the possibility of accumulation and compaction of wear debris at the craters is less and thus the surface becomes relatively rough which results in higher coefficient of friction. At higher applied load, coarser wear debris generated during wear has the greater tendency to be accumulated and compacted at these craters because of combined action of frictional heating, greater degree of material flow and applied load. These would lead to smoother surface and greater degree of sliding action which leads to reduction in coefficient of friction. At higher applied load, irrespective of sliding speed, greater degree of adhesion take place and this leads to delaminating wear especially from the craters because of greater extent of crack generation and weak bonding between compacted wear debris and crater surface. This leads to sharp transition in wear rate due to increase in applied load from 3 to 5 kg. As compared to Al–SiC composites, aluminum syntactic foam exhibits significantly less coefficient of friction which may be attributed to greater abrasive action of protruded SiC particles over the counter surface. The wear rate at lower applied load is noted to be almost same in both the materials. At higher applied load, aluminum syntactic foam exhibits greater wear rate as compared to Al–SiC composite because of the higher strength and less extent of adhesive wear due to presence of rigid and hard SiC particles.
2568
D.P. Mondal et al. / Materials and Design 30 (2009) 2563–2568
References [1] Kiser M, He YM, Zok FW. The mechanical response of ceramic microballon reinforced aluminium matrix composite under compression loading. Acta Mater 1999;47:2685. [2] Kim HS, Plubrai P. Manufacturing and failure mechanism of syntactic foam under compression. Compos Pt A 2004;35:1009. [3] Balch DK, O’Dwyer JG, Davis GR, Cady CM, Gray GT, Dunand DC. Plasticity and damage in aluminium syntactic foams deformed under dynamic and quasi static condition. Mater Sci Eng A 2005;391:408. [4] Palmer RA, Gao K, Doam TM, Green L, Cavallaro G. Pressure infiltrated syntactic foams – process development and mechanical properties. Mater Sci Eng 2007;464:85–92. [5] Wu CML, Han GW. Synthesis of an Al2O3/Al co-continuous composite by reactive melt infiltration. Mater Charact 2007;58:416–22. [6] Mizumoto Masayuki, Ohga Takeshi, Kagawa Akio. Characterization of fiberreinforced metal matrix composites fabricated by low-pressure infiltration process. Mater Sci Eng 2005;413–414:521–6. [7] Rao B Srinivasa, Jayaram V. Pressure less infiltration of Al–Mg based alloys into preforms: mechanisms and phenomenology. Acta Mater Al2O3 2001;49:2373–85. [8] Cook AJ, Werner PS. Pressure infiltration casting of metal matrix composites. Mater Sci Eng A 1991;144:189–206. [9] Ceschini L, Bosi C, Casagrande A, Garagnani GL. Effect of thermal treatment and recycling on the tribological behaviour of an AlSiMg–SiCp composite. Wear 2001;251:1377–85. [10] Singh M, Prasad BK, Mondal DP, Jha AK. Dry sliding wear behaviour of an aluminium alloy–granite particle composite. Tribol Int 2001;34:557–67. [11] Modi OP, Rathod S, Prasad BK, Jha AK, Dixit G. The influence of alumina particle dispersion and test parameter on dry sliding wear behaviour of zinc-based alloy. Tribol Int 2007;40:1137–46. [12] Uyyuru RK, Surappa MK, Brusethaug S. Effect of reinforcement volume fraction and size distribution on the tribological behaviour of Al composite/brake pad tribo-couple. Wear 2006;260:1248–55. [13] Sudrashan, Surappa MK. Dry sliding wear of fly ash particle reinforced A356 Al-composites. Wear 2008;265:349–60. [14] Hamid A Abdulhaqq, Gosh PK, Jain SC, Ray Subrata. Influence of porosity and particles content on dry sliding wear of cast in situ Al(Ti)–Al2O3(TiO2) composite. Wear 2008;265:349–60. [15] Ranganath G, Sharma SC, Krishna M. Dry sliding wear of garnet reinforced zinc/aluminium metal matrix composites. Wear 2001;251:1408–13.
[16] Wang Y, Rainforth WM, Jones H, Lieblich M. Dry wear behaviour and its relation to microstructure of novel 6092 aluminium alloy–Ni3Al powder metallurgy composite. Wear 2001;251:1421–32. [17] Hui Di-Zhou, Chen Jian-Min, Jia Jun-Hong. Investigation of the tribological behaviour of an aluminium alloy with embedded materials. Mater Sci Eng A 2001;302:222–6. [18] Bermudez MD, Nicolas G Martinez, Carrion FJ, Mateo Martinez, Rodriguez JA, et al. Dry and lubricated wear resistance of mechanically alloyed aluminium base sintered composite. Wear 2001;248:178–86. [19] Rodriguez J, Poza P, Garrido MA, Rico A. Dry sliding wear behaviour of aluminium–lithium alloys reinforced with SiC particles. Wear 2007;262: 292–300. [20] Zhang Shaoyang, Wang Fuping. Comparison of friction and wear performance of brake material dry sliding against two aluminium matrix composites reinforced with different SiC particles. J Mater Proc Technol 2007;182:122–7. [21] Ahlatci Hayrettin, Kocer Tolga, Candan Ercan, Cimenoglu Huseyin. Wear behaviour of Al/(Al2O3p + SiCp) hybrid composites. Tribol Int 2006;39:213–20. [22] Cao GH, Wu SQ, Liu JM, Liu ZG. Wear resistance mechanism of an Al–12Si alloy reinforced with aluminisilicate short fibres. Tribol Int 1999;32:721–4. [23] Tjong SC, Wu SQ, Liao HC. Wear behaviour of an Al–12% Si alloy reinforced with low volume fraction of SiC particles. Compos Sci Tech 1998;57:1551–8. [24] Sahin Y, Murphy ST. He effect of sliding speed and microstructure on the dry wear properties of metal matrix composites. Wear 1998;214:98–106. [25] Sun Zhiqiang, Zhang Di, Li Guobin. Effect of dry sliding wear behaviour of silicon particles reinforced aluminium matrix composite. Mater Des 2005;26:454–8. [26] Alpas AT, Embury JD. Sliding and abrasive wear resistance of an aluminium– SiC particle reinforced composite. Scripta Metall Mater 1990:931–5. [27] Ray S, Fishman SG, Dhingra AK, editors. Symposium on advances in cast metal matrix composite: challenges in processing and design. ASM International; 1988. p. 77. [28] Vardavoulias M, Jouannt Tresy C, Jeandin M. Sliding wear behaviour of ceramic particle reinforced high-speed steel obtained by powder metallurgy. Wear 1993;165:141–9. [29] Wang Junan, He Yeng, Danninger Herbert. Influence of porosity on the sliding wear behaviour of sintered Fe–1.5Mo–0.7C steels. J Mater Eng Perf 2003;12:339–44. [30] Mondal DP, Das S, Rao RN, Singh M. Effect of SiC addition and running-in-wear on sliding wear behaviour of Al–Zn–Mg aluminium alloy. Mater Sci Eng A 2005;402:307–19.