Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting

Alexandria Engineering Journal (2017) xxx, xxx–xxx

H O S T E D BY

Alexandria University

Alexandria Engineering Journal www.elsevier.com/locate/aej www.sciencedirect.com

ORIGINAL ARTICLE

Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting I.M. El-Galy *, M.H. Ahmed, B.I. Bassiouny Production Engineering Department, Faculty of Engineering, Alexandria University, Egypt Received 18 November 2016; revised 24 January 2017; accepted 4 March 2017

KEYWORDS Metal matrix composites (MMC); Functionally graded materials (FGM); Centrifugal casting; Aluminium; Silicon carbide; Wear resistance

Abstract The design of metal matrix composites can be enhanced by integrating the concept of functionally graded materials (FGM) to produce engineering materials with tailored contradictory properties that suit multifunctioning components. The present investigation focuses on characterization of functionally graded metal matrix composites (FGMMCs) based on pure aluminium matrix reinforced with different percentages and sizes of SiC particles. The investigated FGMs have been produced by horizontal centrifugal casting process under different conditions. Microstructure investigation, tensile, hardness and wear rate measurements have been correlated with the size and percentage of SiC particles and their distribution/gradient across the thickness of the cast tubes resulting from the used casting parameters. Ó 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Functionally graded structures can be seen in nature, in biotissues of animals, such as bones and teeth, and plants. Dental crowns represent excellent examples of functionally graded structures. They require a high wear resistance outside and a ductile inner structure for reasons of optimal fatigue and brittleness combination [1]. The development of composite materials with graded properties known as functionally graded

* Corresponding author. E-mail addresses: [email protected] (I.M. El-Galy), mhmahmed @alexu.edu.eg (M.H. Ahmed), [email protected] (B.I. Bassiouny). Peer review under responsibility of Faculty of Engineering, Alexandria University.

materials (FGM) has made a revolution in the manufacturing of mechanical parts, especially in the automotive, aviation and biomedical industries [2]. The concept of FGM was first considered in Japan in 1984 during a space vehicle project. That time the aim was to fabricate the body’s material with improved thermal resistance and mechanical properties by gradually changing compositions to withstand severe temperature gap (about 1000 °C) in between the inside and the outside [3]. The need for property distributions is found in a variety of common products that must have mutually exclusive requirements to provide multi-functional characteristics. For example, gears must possess high internal toughness to withstand dynamic loading and superior surface hardness to prevent wear [4]. Similarly, a turbine blade also possesses a property distribution. The blade core must be tough enough to withstand the heavy dynamic loading it is subjected to, while its

http://dx.doi.org/10.1016/j.aej.2017.03.009 1110-0168 Ó 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

2 surface must have a high melting point to be able to withstand high temperatures encountered during operations. As with hardness and toughness, these different properties tend to exclude one another. There are many applications for FGMs in aerospace, power generation, electronics and bioengineering which demand properties that are not achievable in any single material [5]. FGM materials are manufactured using different techniques including powder metallurgy [6,7], vapour deposition [8], solid freeform (SFF) fabrication [9], centrifugal slurry or centrifugal casting [10]. The production of FGM using centrifugal casting involves the addition of hard particles to a molten metal and pouring the mixture into a rotating cylindrical mould. The composite is solidified on the internal wall of the mould under centrifugal force, with gradient of the hard particles depending on the size of particles, mould rotational speed, density and viscosity of the molten metal as well as the rate of solidification [11]. FGMs are manufactured to achieve pre-required gradient in strength, corrosion, wear or high temperature resistance across the wall thickness of the product. Weight saving can be achieved in functionally graded castings compared to casting made entirely from MMCs [12]. Various researchers have carried out the researches on FGM with focus on mathematical modelling and simulation of the particles [13,14] and the production processes [15–17] while others focused on characterization of the mechanical, tribological as well as microstructural characterization of FGMs prepared using different production techniques [18]. Rodriguez-Castro studied the effect of adding 23 lm SiCp on the resulted microstructure and mechanical behaviour of centrifugally cast aluminium alloy A359/SiCp [19]. He investigated the microstructure, hardness, tensile strength and fatigue fracture behaviour of FGMs by varying the percentage of SiCp between 20% and 40% and rotational speed between 700 rpm and 1300 rpm. A maximum increase of nearly 15% in tensile strength and surface hardness could be achieved by the addition of 20% SiCp at 1300 rpm compared to the base A359 alloy. Rajan and Pillai (2008) fabricated discs from A356 alloy by adding 20 wt.% of 23 lm SiCp to the FGMs produced by vertical centrifugal casting. They studied the effects of reinforcement on microstructure, hardness, ultimate tensile strength, and yield strength. Ultimate compressive strength and modulus of elasticity of the samples taken from the FGM disc had been determined [20]. A maximum increase of nearly 26% in hardness and 18% in tensile strength of outer layers could be achieved compared to the base A356 alloy. In 2010 they extended their work and compared the characteristics of FGM based on cast A356 and wrought AW2124 aluminium alloys reinforced with 15 wt.% of 23 lm SiCp [21]. For the A356/15 wt.%SiCp composite, a maximum increase of nearly 38% in outer layers hardness could be achieved compared to the base alloy. On the other hand, only 28% increase in outer layer hardness could be reached in case of composites made of AW2124/15 wt.%SiCp. Vieira and Sequeira (2009) studied the effects of reinforcement of SiCp on sliding wear behaviour of centrifugally cast Al–10Si–4.5Cu–2 Mg alloy reinforced by 10 wt.% of 37 lm SiCp. The centrifugal casting processes were performed at 2000 rpm. By varying wear load and track velocity, they concluded that SiCp reinforced material has better wear resistance

I.M. El-Galy et al. with increase of SiCp content [22]. Brinell hardness of 160 BHN could be reached at outer layers, while the hardness of the inner layers recorded 105 BHN. Wear loss in the order of 0.01 mg was recorded at outer layers, compared to 10 mg weight loss in the base non-reinforced alloy. Although this investigation resulted in very high hardness and wear resistance levels, it did not show the negative influences on tensile and ductility properties. Vikas and Maiya (2014) fabricated functionally graded rings using vertical centrifugal casting process. The composite rings were made of AA6061 reinforced with 10 wt.% SiCp of 23 lm particle size. They investigated the microstructure and evaluated the hardness and wear rate behaviour of the FGM rings under different loading levels. The measured hardness levels at outer layers reached a maximum of 80 BHN. This represents 38% higher than the hardness measured at the inner layers of the cast rings [23]. Jayakumar and Rajan (2016) characterized functionally graded metal matrix composites (FGMMC) made of A319 alloy reinforced with 23 lm SiCp. Rings of FGMs with 10 and 15 wt.% SiCp were produced by vertical centrifugal casting process. The researchers evaluated the mechanical characteristics in addition to the microstructure, coefficient of thermal expansion and wear behaviour of FGMs [24]. The authors referred to the presence of an outer chill zone, but the maximum hardness was measured at the adjacent inner zones towards the centre. Maximum Brinell hardness (65 BHN) has been measured in composites with 10 wt.%SiCp at the outer zones, compared to 45 BHN at the inner zones, whereas the maximum strength has been measured in case of samples taken from the outer zone only for composites containing 15 wt.%SiCp. Recently, Radhika and Raghu (2016) investigated the behaviour of FGMs produced from A319 alloy reinforced with B4C, SiC, Al2O3 or TiB2. They compared the microstructure, hardness, tensile strength and wear rate [25]. The maximum tensile strength could be achieved by adding 10 wt.%TiB of 10 lm particle size, while the maximum outer zone hardness is realized by adding 15 wt.%SiC of 23 lm particles. The given review shows that there is a focus on using Al-Si alloys (e.g. A356), which possess higher strength and hardness, but lower melting temperature and viscosity compared to pure aluminium. This may facilitate the production of the FGM by expanding the window available for pouring and results in higher strength composites. In addition, the existence of Si within the melt prevents the dissolution of SiCp [26] and the formation of the hydrophilic and brittle Al4C3 [27]. The formation of this carbide at the matrix/SiCp interface reduces the interfacial strength as well as the fracture energy and increases the corrosion sensitivity [28]. However, the lower melting temperature of Al-Si alloys makes the FGM more vulnerable to softening in high temperature applications. In addition, the cost of the alloy is higher than that of pure aluminium, which introduces involvement of economic consideration to the FGM cost. In this study, the investigations were carried out using commercially pure aluminium to find out a suitable procedure to produce centrifugally cast FGM products in the tight available pouring time. The effect of higher pouring temperature and hence the lower molten metal viscosity on the distribution of the SiC particles was investigated. Moreover, the large temperature gradient between the molten metal and the mould should

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

Characterization of functionally graded Al-SiCp metal matrix composites prevent the formation of aluminium carbides by reducing the free energy required for their formation. On the other side, most research work focused on small sized particles (16–63 lm). The larger sized particles are usually used for surface hardening because of the large centrifugal forces due to the particle mass. In the present work, the effects of large particles (500 lm) have been studied and compared to small sized particles (16 and 23 lm). The large sized particles might be suitable to overcome the higher viscosity of pure aluminium, so that a functionally graded structure can be obtained. Large particles also assure higher inertia to overcome the progressively advancing solidification interface encountered due to the high temperature gradient. This research work aims at fabricating functionally graded tube made of pure aluminium reinforced with SiC particles by horizontal centrifugal casting. Production parameters include size and weight fraction of SiC particles, rotational speed and pouring rate. Effects of SiCp addition on microstructure, hardness, tensile strength, ductility and wear rate are also included. 2. Tests and analysis techniques 2.1. Fabrication of pure Al-SiC FGMs by centrifugal casting The centrifugal casting machine, whose schematic is shown in Fig. 1, was designed and manufactured to perform the required experiments. The machine is equipped with a variable frequency drive (VFD) to control the mould rotational speed in the range of 500–2500 rpm, as well as, second VFD to control the axial speed of the feeder tube in the range 5–50 mm/s. Commercially pure aluminium (99.97%, 25 BHN, 80 MPa UTS) has been used as a matrix material. Reinforcement was done by adding SiC particles. FGMMCs were manufactured with different weight fractions of SiCp by centrifugal casting technique. The FGM production procedure includes the following: 1. Melting of the base metal in a graphite crucible at 670 °C and heating to the pouring temperature of 725 °C. 2. SiCp powder is added at different weight percentages (0, 2.5, 5, 7.5, 10 and 15 wt.%) of the molten metal weight. Three different SiCp sizes with average particle size of

Figure 1

3

16 lm, 23 lm or 500 lm were used. Fig. 2 shows a relative comparison of the used SiCp particles. The 500 lm particles exhibit more uniform spherical shape and size, compared to both of the 16 lm and 23 lm particles. The smaller Lely particle irregularity is due their manufacturing through crushing process. 3. The SiC particles were mixed with stirrer at a speed of 100 rpm 4. The mixture is poured into a centrifugal casting machine which has been adjusted to different rotation speeds (800, 900 or 1000 rpm) to produce FGM tube with outer diameter of 230 mm, wall thickness of 12 mm and length of 180 mm. 2.2. Characterization of material 2.2.1. Investigation of microstructure Metallographic samples were sectioned from the cast FGM tubes through the wall thickness. The samples have been polished and etched with 0.5% diluted Hydrofluoric acid (HF). Microscopic examination has been performed using Axiovert 25 CA compound optical microscope. The difference in distribution of SiC particles in the aluminium matrix has been determined. 2.2.2. Evaluation of mechanical properties Different samples have been prepared to perform mechanical testing of the resulting FGM structures. Fig. 3 shows the locations of the cut samples for use in tensile, hardness and wear testing. The tensile test specimen and a schematic drawing with the used dimensions are shown in Fig. 4. The samples were cut on a vertical milling machine and were finished by fine emery paper (grit 600). The specimens were machined with the complete tube thickness to investigate the effect of particle distribution through the thickness for each of the studied SiCp weight percentages. The tensile strength and the elongation at fracture were determined in a tensile test. The test was carried out according to ISO 1608:1995 (mechanical testing of metalstensile testing of materials) on TUN-400 tensile testing machine. Brinell hardness test has been conducted on specimens of varying weight fraction of SiC particles. Hardness testing machine (Model MRB-250) has been used with a load of 62.5 N and a steel ball indenter of 5 mm diameter.

Schematic of the horizontal centrifugal casting machine.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

4

I.M. El-Galy et al.

Figure 2

Comparison of the used particle shapes and sizes.

grinding of the specimens. The rotating disc material is made of EN-31 steel (63 HRC). The pin was loaded with 24 N and is allowed to rub against the rotating disc on a 170 mm diameter track at 900 rpm. The dry sliding wear rate was has been evaluated by weight loss of the specimens. The accuracy of the used weighing device is 1 mg. 3. Results and discussion 3.1. Investigation of microstructure

Figure 3

Locations of the cut samples.

The dry sliding wear tests were performed on pin-on-disc apparatus. The samples were cut from the FGM tube to 20
20
10 mm, and the chill layer was removed by flat

Figure 4

Four distinct zones could be identified in each of the 12 mm thick sample: chill, outer, concentration transition and inner zones, respectively. Investigation of the samples shows the graded distribution of SiC particles in the matrix with the highest SiCp concentrations at the outer zone. The limits of each zone are defined according to the gradient of SiCp concentration as it will be shown later in Section 2.2. Figs. 4–6 show the microstructure of samples taken through the thickness of the produced functionally graded tubes. Fig. 5 shows the different four zones characterizing these

Example of tensile testing specimen.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

Characterization of functionally graded Al-SiCp metal matrix composites

5

Pure Al +10 wt.% SiCp (1000 rpm, 23 m, 16 mm/s feed)

a) Chill to Outer Zone

b) Concentration Transition Zone

c) Inner Zone

Microstructure of FGM with 10 wt.% SiCp of 23 lm.

Figure 5

(Outer Zones, 1000 rpm, 16 m, 16 mm/s feed)

a) 2.5% SiCp Figure 6

b) 10% SiCp

c) 15% SiCp

Microstructure of FGM with different weight fractions of SiCp of 16 lm.

Pure Al +10 wt.% SiCp (1000 rpm, 500 m, 16 mm/s feed)

a) Chill Zone and Outer Zone Figure 7

b) Transition Zone

c) Porosity in inner Zone

Microstructure of FGM with 10 wt.% SiCp of 500 lm.

centrifugally cast tubes. Each of the three images represents the typical distribution within a zone of 4 mm through the thickness. Fig. 5a represents the chill and outer zones, while Fig. 5b represents the concentration transition zone, and Fig. 5c represents the start of the inner zone close to the end of the transition zone. The rapidly cooled chill zone (outermost 0.5:1 mm) could be observed in FGM composites with particle sizes of both 16 and 23 lm for all weight fractions of SiCp. It is characterized by relatively low percentage of particles, especially in the outmost 0.5 mm. In case of 500 lm particles, the chill zone is much thinner and the percentage of filling with particles is comparatively higher (Fig. 7a). Next to the chill zone lies the outer zone. For all examined weight fractions (Fig. 6) and particle sizes (Figs. 5a, 6b and 7a),

the outer zone is enriched with the highest concentration of SiCp because of the pushing of particle from inner periphery to the outer periphery where the solidification happens slowly compared to the chill zone. This could be confirmed by the evaluated concentrations in the different zones, which have been measured using the apparent area of the particles using image analysis technique as given in Figs. 8–10. The outer zones of specimens containing 500 lm SiCp (Fig. 7), show higher concentration of particles than the outer zones in tubes with 16 lm and 23 lm particles (Figs. 5 and 6) because the effect of centrifugal force in case of the larger particles is much higher. Fig. 5b represents the concentration transition zone of the 23 lm particles. Scanning the transition zone, shows a gradual decrease in percentage of particles towards the inner zone.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

6

I.M. El-Galy et al. 23 m

50

500 m

Process condions 1000 RPM 2.5% SiC 28 mm/sec feed

40 30

Conc.Transion Zone

10

Inner Zone Outer Zone

0 1

2

3

4

5

6

7

8

40

Conc. Transion Zone

30 20 10

0

Inner Zone

Outer Zone

1

2

3

4

5

6

7

8

9 10 11 12

Distance from Outer Surface (mm)

a) 2.5 wt.% SiC

b) 5 wt.% Sic

23 m

500 m

Process condions 1000 rpm 10% SiCp 28 mm/sec feed

50 40

Conc. Transion Zone

30

10

Inner Zone

Outer Zone

16 m

23 m

500 m

Process condions 1000 RPM 15% SiC 28 mm/sec feed

50 40 30 Chill Zone

20

Concentraon of SiC (area %)

60 16 m

Chill Zone

Concentraon of SiC (area %)

500 m

Distance from Outer Surface (mm)

0

20 10

Inner Zone

Transion Zone

Outer Zone

0 0

1

2

3

4

5

6

7

8

9 10 11 12

0

1

2

3

4

5

6

7

8

9 10 11 12

Distance from Outer Surface (mm)

Distance from Outer Surface (mm)

c) 10wt.% SiC

d) 15 wt.% Sic

Figure 8

Effect of particle size on distribution of SiCp at different weight fractions. 70

23 m, 16 mm/s 23 m, 28 mm/s 500 m, 16 mm/s 500 m, 28 mm/s

60 50

Process condions 1000 RPM 10 % SiC

40 30 20 10

Outer Zone

Inner Zone

50

30 20 10 0

0

1

2

3

4

5

6

7

8

9

10 11 12

Distance from Outer Surface (mm) Effect of rotational speed on concentration of SiCp.

Samples with 23 lm exhibit better gradient distribution than those observed in case of 16 and 500 lm particles. It is noticed that there are some clusters of particles in the microstructure when compared to 16 and 500 lm. In case of 500 lm particles, it has been observed that there is minor difference in the distribution and filling when the

Process condions 1000 RPM 10 % SiC

40

Conc. Transion Zone

Chill Zone

Chill Zone

Conc. Transion Zone

23 m, 16 mm/s 23 m, 28 mm/s 500 m, 16 mm/s 500 m, 28 mm/s

60

Concentraon of SiC (area %)

70

Concentraon of SiC (area %)

50

9 10 11 12

60

Figure 9

23 m

Process condions 1000 rpm 5% SiC 28 mm/sec Feed

0 0

0

16 m

Chill Zone

20

60

Concentraon of SiC (area %)

16 m

Chill Zone

Concentraon of SiC (area %)

60

0

Outer Zone 1

2

3

Inner Zone

4

5

6

7

8

9

10 11 12

Distance from Outer Surface (mm) Figure 10 Effect of feed on distribution of SiCp for different particle sizes.

transition zone (Fig. 7b) is compared to the outer zone (Fig. 7a). In addition, the extent of both zones is narrower than observed in small-sized particles. The wide inner zone in this case starts at position of 5–8 mm from the outer periphery depending on the other processing parameters.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

Characterization of functionally graded Al-SiCp metal matrix composites The inner zone in this Fig. 7-c exhibits the presence of porosities due to the settling of the low-density inclusions and agglomerations due to the lack of centrifugal force nearer to the centre of the mould which affects the solidification rate. For the inner zone, the gradient of 23 lm particles is still more efficient than the 16 lm while in the 500 lm there is a very low density of SiC particles. The results obtained for the SiCp distribution resembles the distributions obtained by other researchers [20,21]; however, some discrepancies are noticed in the concentrations due to the different process conditions (higher rotational speeds or wt.% of SiCp). 3.2. Effect of the SiCp size on the distribution of particles through thickness 3.2.1. Particle analysis Image processing technique has been applied with the aid of Image J software program to analyse the distribution and the number of silicon carbide particles in the different zones of the cast tubes. Automatic tools of particles detection have been applied to define edges, adjust contrast, count the particles and calculate relative concentration of SiCp in the acquired image. The image analysis results are represented in Figs. 8–10. Each of the data points shown on the graphs represents the average concentration of the SiCp in 1 mm thick zone. Fig. 8 shows the concentration of SiCp through the different zones of FGM tube reinforced with different weight fractions of 2.5, 5, 10 and 15 wt.% SiCp. For the zone range 0 to 1 mm (the chill zone), the average concentration measured as percentage of area increases from 18% to 37% with the increase in weight fraction of SiCp from 2.5% to 15% in case of 16 lm particles. The same trend is observed for the 23 lm particles but with higher percentage of area (1–5% max deviation). Nearly 10% higher concentration of SiCp is observed in case of 500 lm particles. The outer zones exhibit the maximum percentage of SiCp in the second millimetre measured from the outer periphery of the produced tubes. These maximum concentrations are proportional to the weight fraction of SiCp as well as to the particle sizes. However, the average values are much higher in case of 500 lm particles. The measured average concentration varies between 24% and 44% for the 16 lm particles while the variation ranges from 44% to 55% in case of 500 lm. This increase is followed by a gradual decrease in the concentrations towards the inner diameter. The decrease in the concentration rate is much higher in case of the 500 lm than the other two particle sizes. The total drop from the maximum concentration - start of outer zone - to the minimum concentration - start of the inner zone with nearly 0% SiCp - can be divided into two zones with different two gradients: the outer zone with the lower gradient, and the transition zone with higher gradient. The gradient of the SiCp concentration can be evaluated by the %SiCp reduction per mm. For example, in case of 10 wt.% SiC, the total gradients from the maximum to minimum concentrations were calculated as 6%, 6.42% and 14% SiCp/mm for 16, 23 and 500 lm particles, respectively. The distribution of the SiCp at different weight fractions and small particle sizes is comparable to the results obtained

7

by [20–22,17]. However, the researchers did not provide any results for large particle sizes. In addition, the difference of the properties change with different gradients has not been discussed even if the presented graphs ensure the existence of effect [17]. 3.2.2. Effect of the mould rotational speed on distribution of particles Fig. 9 shows the relationship between concentration of SiCp and rotational speed for 10 wt.% SiCp FGM tube. With increasing rotational speed, the maximum weight fraction of SiC particles obtained at the outer periphery is increased due to higher centrifugal force by increasing the mass of particles. For the chill zone, the concentration for specimens manufactured at 1000, 900 and 800 rpm is 42%, 38% and 36%, respectively. The concentrations of the SiC particles in the zone range from 1 to 2 mm increased to its maximum value of 45%, 41% and 37% for the 1000, 900 and 800 rpm, respectively followed by a gradual decrease in the direction of inner diameter. The gradient of the concentration change is the same in the outer zone for all used rotational speeds. However, there is a large change in gradient for the concentration transition zone. The total gradient in the concentration rate is much higher in case of the 1000 rpm compared to the other two speeds. The total gradient of the SiCp concentration at the 800, 900 and 1000 rpm is 5%, 8.5% and 13.5% SiCp/mm, respectively. The effect of changing the centrifugal casting speed has been found to be comparable to the results provided by [12,15]. However, they obtained higher concentrations at the outer zones because they used higher speeds in the range 1300–2000 rpm. 3.2.3. Effect of the pouring mechanism feed on distribution of particles Fig. 10 shows the relationship between concentration of SiCp and pouring feed rates in case of 10 wt.% SiCp, with particle grain sizes of 23 lm and 500 lm as an example. The produced FGM tubes contain a maximum of 45% SiCp in case of 23 lm, and 58% in case of 500 lm at a feed rate of 28 mm/s. This is followed by a gradual reduction to lower levels. The decrease in the concentration rate is much higher in case of the low feed speed (16 mm/s) than at high feed speed (28 mm/s). With increasing the feed, the maximum concentration of SiCp obtained at the outer zones increased due to higher centrifugal force by increasing mass of particles. The gradient of the concentration change is the same in the outer zone for all used axial feed speeds. However, there is a large change in gradient for the concentration transition zone. The gradient of the SiCp concentration in the transition zone is comparable for both grain sizes and feeds. Comparison of the influences caused by studied parameter shows that, the effects of rotational and particle grain size on the SiCp concentration gradient are more than the effects caused by changing the axial feed speed. 3.3. Effect of reinforcement on hardness Fig. 11 shows the variations of the Brinell hardness values for the FGM tube samples taken from different layers through the thickness corresponding to the mid-points of the chill, the outer, the concentration transition and the inner zones.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

8

I.M. El-Galy et al.

3.4. Effect of SiCp distribution on ultimate tensile strength and ductility Fig. 13 shows the ultimate tensile strength of FGM tube versus SiCp weight fraction. The addition of SiCp can apparently enhance the ultimate tensile strength of FGM tube. At 5 wt. % SiCp, the ultimate tensile strength reached 115, 106 and 100 MPa for 16, 23 and 500 lm, respectively. In case of 15 wt.% SiCp, the tensile strength increased to a maximum value of 146, 144 and 130 MPa for 16, 23 and 500 lm, respectively. The strength increase with increasing the weight fraction is the most effective in case of 16 and 23 lm than for 500 lm.

Brinell Hardness

48 44 40

Chill Zone

Outer Zone

Transion Zone

Inner Zone

16 m

55

36 32

500 m

Process Condions 23 m, 1000 rpm 16 mm/sec feed 62.5 Kg load Ball Diameter 5 mm

52 49 46 43 40 37 34 31 28 22

0

2.5

5

7.5

10

15

Weight Fracon (%SiCp) Figure 12 Effect of SiCp weight fraction on Brinell hardness of outer zone.

The presence of high concentration of SiCp provides higher strength near the region towards the outer periphery of the centrifugally cast tubes. However, very high compaction of particles as in case of 500 lm may lead to smaller amount of matrix materials between the particles and hence lower binding. Fig. 14 shows the ductility of FGM tube versus SiCp weight fraction. The maximum elongation of tested FGM specimen shows an exponential decrease with increasing the percentage of SiC particles. Ductility of 27% is realized at 5 wt.% SiCp, whereas a maximum elongation of 18% is achieved at 15 wt.% SiCp for 500 lm particles. Comparison of the obtained hardness and tensile results to the results of the work done by [20,23,24] shows the same trend and percentage increase for small sized particles. However, the higher strength obtained by other researchers is mainly due to the use of aluminium alloys with higher initial hardness and ultimate tensile strength.

160

140

Process Condions 23 m, 1000 rpm, 16 mm/s feed 62.5 Kg load Ball Diameter 5 mm

23 m

25

Tensile Strength (MPa)

52

58

Brinell Hardness of Outer Zone

For tubes made of pure metal (without particles), the maximum hardness has been measured at the chilled zone which is in direct contact with the mould. The hardness measured at 3, 7 and 10 mm from the outer surface exhibited an average hardness of 25 BHN, with a little decrease towards the inner zones. For FGM tubes, the results show that the hardness increases with increasing weight fraction of SiC particles from 2.5 wt.% to 15 wt.%. However, the chilled layer exhibits a lower increase in hardness compared with the outer and concentration transition zones. The increase of hardness values measured in the outer and concentration transition zones is comparable. Due to the very low weight fraction of particles in the inner zone, the hardness values did not show any increase compared to the values determined for the base metal. Fig. 12 shows the variations of the Brinell hardness values at the outer zones of the FGM tube samples produced by adding SiCp with different sizes and percentages. The hardness increases with decreasing particle size of SiC particles. At 2.5 wt.% SiCp, the measured Brinell hardness values were 34, 31 and 29 for the 16 lm, 23 lm and 500 lm, respectively. At 10 wt.% SiCp, the Brinell hardness values were 49, 46 and 42 for the 16 lm, 23 lm and 500 lm, respectively. It should be noticed that the increase in the overall hardness with decreasing the particle size occurs although the %SiC particles at the chill and outer zones are higher in concentration compared to the smaller particles (Fig. 8). This increase may be due to the backing effect caused by the smaller inter-distances among the particles of smaller sizes.

16 m

23 m

500 m

Process Condions 1000 rpm 16 mm/ sec Feed

120

100

80

28 24

0

2.5

5

7.5

10

60

15

Weight Fracon (%SiCp) Figure 11 Effect of SiCp weight fraction on Brinell hardness at different zones.

0

5

10

15

Weight Fracon (% SiCp)

Figure 13 strength.

Effect of SiCp weight fraction on ultimate tensile

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

Characterization of functionally graded Al-SiCp metal matrix composites 3.5. Effect of SiCp distribution on wear

In this section, results of the wear test for samples prepared from the outer, transition and inner zones for different SiCp weight fractions are presented. One millimetre of both surface layers, on inner and outer sides, has been removed by machining prior to wear testing to prepare flat samples. Examples of the wear loss for 5, 10 and 15 wt.% SiCp are shown in Fig. 15. The highest wear resistance is achieved at the outer zones for all reinforced FGM structures including 2.5 and 7.5 wt.% SiCp. The low weight loss in the outer zone can be attributed to the high content of reinforcing particles. It could be shown that the weight loss decreased linearly with increasing the weight fraction of the SiCp in both outer and concentration transition zones. However, a lower effect could be realized by increasing the weight fraction from 10% to 15%. 3.5.2. Effect of weight fraction of SiCp on wear of outer zone Fig. 16 shows the cumulative weight loss of the outer zone of FGM specimens with time. The curves represent the progress of wear for samples with different weight fractions. After addition of reinforcing material, the sliding wear of the outer zones of the tube is significantly decreased. The progress of wear loss over time can be described by a logarithmic function for all tested weight fractions. The maximum increase in wear resistance could be achieved by adding between 5 wt.% and 10 wt.% SiCp. Further increase in the particles up to 15% resulted in lower increase in wear resistance. Macro examination of the FGMMC shows better surface quality and less evidence of wear compared to pure aluminium as shown in Fig. 17. The obtained wear results in this study show the similar trend when compared to the results obtained by [22–25]. 3.5.3. The relation between wear loss and processing parameters Multiple regression analysis has been performed based on the results of the wear test carried out on the outer zones in attempt to determine a time-based empirical relationship between the wear loss and SiCp weight fraction. The estimated wear loss progress is determined under testing load of 24 New-

16 m

23 m

500 m

Process Condions 1000 RPM 16 mm/ sec Feed

Duclity (%)

35

25

15

5

0

5

10

15

Weight Fracon (%SiCp) Figure 14

Effect of SiCp weight fraction on ductility.

Weight Loss in Outer Zone (grams)

0.25

3.5.1. The weight loss at different zones

45

9 5 % SiCp

10 % SiCp

15 % SiCp

Process Condions 1000 rpm 23 m 16 mm/sec feed Test load 24 N Duraon 1 min

0.2

0.15

0.1

0.05

0

Outer Zone

Inner Zone Zone

Figure 15 Weight loss at different zones with increased SiCp weight fraction.

ton at wear disc speed of 900 rpm. Eq. (1) estimates the wear loss progress for tubes produced at mould rotational speed of 1000 rpm, SiC particle size of 23 lm, pouring temperature of 725 °C and feed of 16 mm/s. The given equation is valid for the tested particle size and range of SiCp weight fractions. Further investigations are being carried out to determine the relative influence of each of the process parameters (particle size, pouring temperature and rotational speeds to reach a generalized relationship). Weight LossðgmÞ ¼ 0:0833  e0:008Wf lnðtÞ þ 0:1582e0:0085Wf

ð1Þ

where t is the time in minutes and Wf is the weight fraction 4. Conclusion Functionally graded metal matrix composites made of commercially pure aluminium reinforced with SiCp has been fabricated successfully through horizontal centrifugal casting technique. Different weight fractions (0%, 2.5%, 5%, 7.5%, 10% and 15%) of SiCp with three different particle sizes of 16, 23 and 500 lm have been investigated. Three rotational speeds of 800, 900 or 1000 rpm were used. Two controlled linear speeds of 16 or 28 mm/s have been used for feeding the metal along the tube axis. Investigation of microstructure reveals that the concentrations of the SiC particles in the outer zone of the cast tubes reach its maximum value followed by a gradual decrease in the direction of inner diameter. In case of large particle sizes and higher rotational speeds, all fabricated tubes revealed high concentration of reinforcing particles in the outer zone due to higher centrifugal force and particle mass. In case of particles with 16 or 23 lm, the outer zone contained less concentrations of SiCp due to lower centrifugal forces realized by smaller and lighter particles. At higher axial speeds of feeding tube, the concentration at the outer zones is higher than that obtained at lower feed speeds. However, the gradient of concentration reduction in lower feed conditions is lower than the gradient obtained at

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

10

I.M. El-Galy et al. 0 % SiC

2.5% SiC

5 % SiC

7.5% SiC

10% SiC

15% SiC

Wear Loss of Outer Zone (gm)

0.25

increased wear resistance

Process condions 1000 rpm, 10% SiC, 23 μm 16 mm/sec Feed, Test Load 24 N

0.20

y = 0.0697ln(x) + 0.1496 y = 0.0687ln(x) + 0.1324 y = 0.0669ln(x) + 0.1117 y = 0.0669ln(x) + 0.1117

0.15

y = 0.0534ln(x) + 0.0896

0.10

y = 0.0367ln(x) + 0.0634 0.05

0.00

0

0.5

1

1.5

2

2.5

3

Time (min) Figure 16

The relation between weight fraction and weight loss with time.

Pure Al as cast

0% SiCp

5% SiCp

7.5 % SiCp

10 % SiCp

15 % SiCp

Figure 17

Examples of wear encountered in outer zone of FGM tube after 1 min.

higher feeds and results in smoother change of properties across the thickness. Brinell hardness measurements reveal that high hardness is obtained on the outer zone of all tested FGMs compared to that measured in chill, concentration transition and inner zones. The hardness obtained in case of the smallest particle sizes is the highest among all tested particle sizes. It should be also noticed that the increase in SiCp weight fraction resulted in a proportional increase in outer zone hardness. The rate of increase decreases slightly beyond 10 wt.% SiCp. By increasing the weight fraction of SiCp an increase in tensile strength of FGMs samples cut through the whole tube thickness could be measured, while ductility has decreased. The increase in strength and decrease in ductility have been determined in all cases regardless of the size and distribution of SiCp through the thickness. The ultimate tensile strength has been found to be proportional to the percentage of SiCp and inversely proportional to the size of the particles. A linear

increase in tensile strength is observed up to 10 wt.% SiCp; the increase rate is lower afterwards up to 15 wt.% SiCp. This means that a level of saturation may be reached by adding more SiCp. Further work is needed to investigate the effect of adding more SiCp. Investigation of the wear resistance of the FGM reinforced with 23 lm particles shows that the highest wear resistance is achieved on the outer zones in all investigated tubes regardless of the percentage of SiCp weight fraction. Maximum improvement of wear resistance could be achieved in the range 7.5–10 wt.% SiCp. Further experimental work is required to investigate the effect by adding SiCp of different particle sizes.

Acknowledgment The authors would like to express their gratitude to the technical staff of the Production Engineering Department for their

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009

Characterization of functionally graded Al-SiCp metal matrix composites highly appreciated assistance. The authors would like to thank the Faculty of Engineering, Alexandria University, for the provided financial support. Special thanks to LORD INTERNATIONAL Co. for the technical assistance within the framework of this project.

[15]

References

[16]

[1] Gupta, M. Ankit, Recent development in modeling and analysis of functionally graded materials and structures, Progr. Aerosp. Sci. 79 (2015) 1–5. [2] Y. Miyamoto, W.A. Kaysser, B.H. Rabin, A. Kawasaki, Renee` G. Ford (Eds.), Functionally Graded Materials Design, Processing and Applications, Springer US, Boston, MA, 1999, http://dx.doi.org/10.1007/978-1-4615-5301-4. [3] R. Kumar, D.C.N. Chandrappa, Synthesis and characterization of Al-SiC functionally graded material composites using powder metallurgy techniques, Int. J. Innov. Res. Sci., Eng. Technol. 3 (8) (2014) 15464–15471. [4] F. Erdemir, A. Canakci, T. Varol, Microstructural characterization and mechanical properties of functionally graded Al2024/SiC composites prepared by powder metallurgy techniques, Trans. Nonferrous Metals 25 (2015) 3569–3577. [5] P. Muller, P. Mognol, J.-Y. Hascoet, Modeling and control of a direct laser powder deposition process for functionally graded materials (FGM) parts manufacturing, J. Mater. Process. Technol. 213 (2013) 685–692. [6] D.W. Hutmacher, M. Sittinger, a.M.V. Risbud, Scaffold-based tissue engineering: rationale for computer-aided design and solid freeform fabrication systems, Trends Biotechnol. 22 (7) (2004) 354–362. [7] Y. Watanabe, Hisashi, Review Fabrication of Functionally Graded Materials under a Centrifugal Force, in: Chapter 7 in Nanocomposites with Unique Properties and Applications in Medicine and Industry, Ed. John Cuppoletti, 2011, pp. 134–150, ISBN 978-953-307-351-4. [8] B. Kieback, A. Neubrand, H. Riedel, Processing techniques for functionally graded materials, Mater. Sci. Eng. A 362 (2003) 81– 105. [9] M.M. Gasik, Functionally graded materials: bulk processing techniques, Int. J. Mater. Product Technol. 39 (2010) 20–29. [10] J. Singh, Amit Chauhan, Characterization of hybrid aluminum matrix composites for advanced applications - a review, J. Mater. Res. Technol. (2015) 2–11. [11] R.M. Mahamood, E.T.A. Member, Functionally Graded Material: An Overview, in: Proceedings of the World Congress on Engineering, London, U.K., July 4–6, 2012. [12] B.C. Pai, T.P.D. Rajan, Developments in processing of functionally gradient metals and metal-ceramic composites: a review, The Chinese Society for Metals and Springer-Verlag, Berlin Heidelberg 27 (5) (2014) 825–838. [13] J. Gao, C. Wang, Modeling the solidification of functionally graded materials by centrifugal casting, Mater. Sci. Eng. 292 (2) (2000) 207–215. [14] A. Velhinho, P. Sequeira, Al/SiCp functionally graded metalmatrix composites produced by centrifugal casting: effect of

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

11

particle grain size on reinforcement distribution, Mater. Sci. Forum. (2003) 257–262. P.E.J. Babu, T. Rajan, S. Savithri, U. Pillai, B. Pai, Theoretical Analysis and Computer Simulation of the Particle Gradient Distribution in a Centrifugally Cast Functionally Gradient Materials, in: Int. Symp. of Research Students on Mat. Sc. Eng., India, Dec., 2004. G.H. Rahimi, M.Z. Nejad, Deformations and stresses in rotating fgm pressurized thick hollow tube under thermal load, Sci. Res. Essay 4 (3) (March 2009) 131–140. P. Abdul Samad, K. Sandeep, Mathematical modeling and computer simulation of particle gradient distribution in a vertical centrifugally cast functionally gradient composite, Int. J. Innov. Res. Sci., Eng. Technol. 3 (7) (July 2014) 441–447. J. Sobczak, L. Drenchev, Metallic functionally graded materials: a specific class of advanced, J. Mater. Sci. Technol. 29 (4) (2013) 297–316. R. Rodriguez-Castro, Processing, Microstructure and Mechanics of Functionally Graded Al A359/SiCp Composite, Ph.D. Thesis, State University of New York at Buffalo, November 2000, 195 p. T.P.D. Rajan, R.M. Pillai, B.C. Pai, Centrifugal casting of functionally graded aluminium matrix composite components, Int. J. Cast Metals Res. 21 (1-4) (2008). T. Rajan, R. Pillai, B. Pai, Characterization of centrifugal cast functionally graded aluminum-silicon carbide metal matrix composites, Mater. Charact. 61 (10) (2010) 923–928. A. Vieira, P. Sequeira, J. Gomes, L. Rocha, Dry sliding wear of Al Alloy/SiCp functionally graded composites: influence of processing conditions, Wear 267 (1–4) (2009) 585–592. R.S. Vikas, M.U. Maiya, E. Jayakumar, T.P.D. Rajan, P.B.C, Nagaraja, Processing and Characterization of SiCp Reinforced Functionally Graded AA 6061 Aluminium Metal Matrix Composites, Int. J. Adv. Mech. Aeronaut. Eng., vol. 1, 2014, pp. 61–65. E. Jayakumar, J.C. Jacob, T. Rajan, M. Joseph, B. Pai, Processing and Characterization of Functionally Graded Aluminum (A319)-SiCp Metallic Composites by Centrifugal Casting Technique, Metall. Mater. Trans. 47 (8) (2016) 4306– 4315. N. Radhika, R. Raghu, Development of functionally graded aluminium composites using centrifugal casting and influence of reinforcements on mechanical and wear properties, Trans. Nonferrous Met. Soc. China 26 (4) (2016) 905–916. Q. Hu, H. Zhao, F. Li, Microstructures and properties of SiC particles reinforced aluminum-matrix composites fabricated by vacuum-assisted high pressure die casting, Mater. Sci. Eng. A (October) (2016) 1–8. M. Besterci, F. Dobesˇ , K. Su¨lleiova´, O. Velgosova´, Mechanical Properties of Al-Al4C3 Composite Produced by Mechanical Alloying, vol. 1, no. 2, 2013, pp. 31–38. M.E.Z.D.E. Salazar, L. Gil, M.D. Escalera, J.L. Baldonedo, A. Uren, Scanning and transmission electron microscopy study of the microstructural changes occurring in aluminium matrix composites reinforced with SiC particles during casting and welding : interface reactions, vol. 196, no. June 1998, pp. 124– 136, 1999.

Please cite this article in press as: I.M. El-Galy et al., Characterization of functionally graded Al-SiCp metal matrix composites manufactured by centrifugal casting, Alexandria Eng. J. (2017), http://dx.doi.org/10.1016/j.aej.2017.03.009