Effects of particle size and distribution on the mechanical properties of SiC reinforced Al–Cu alloy composites
Zhangwei Wang,Min Song,Chao Sun,Yuehui He
DOI: https://doi.org/10.1016/j.msea.2010.11.028
2011-01-01
Abstract:This paper studied the combined effects of particle size and distribution on the mechanical properties of the SiC particle reinforced Al–Cu alloy composites. It has been shown that small ratio between matrix/reinforcement particle sizes resulted in more uniform distribution of the SiC particles in the matrix. The SiC particles distributed more uniformly in the matrix with increasing in mixing time. It has also been shown that homogenous distribution of the SiC particles resulted in higher yield strength, ultimate tensile strength and elongation. Yield strength and ultimate tensile strength of the composite reinforced by 4.7 μm sized SiC particles are higher than those of composite reinforced by 77 μm sized SiC particles, while the elongation shows opposite trend with yield strength and ultimate tensile strength. Fracture surface observations showed that the dominant fracture mechanism of the composites with small SiC particle size (4.7 μm) is ductile fracture of the matrix, accompanied by the “pull-out” of the particles from the matrix, while the dominant fracture mechanism of the composites with large SiC particle size (77 μm) is ductile fracture of the matrix, accompanied by the SiC particle fracture. Keywords Composites Mechanical properties Microstructures Particle size Particle distribution 1 Introduction SiC reinforced aluminum alloy composites are the typical candidates for engineering applications in aerospace, military and civil manufacturing industries due to their enhanced mechanical properties over the corresponding aluminum alloys such as high strength, modulus, wear resistance and fatigue resistance [1–4] . Among different shaped reinforcements, the composites reinforced with particulates offer relatively isotropic mechanical properties compared to the composites reinforced with short fibers or whiskers, and can be produced using conventional metal manufacturing process with low cost [5,6] . During the past two decades a large number of the investigations have been carried out to reveal the strengthening mechanisms of the metal matrix composites. Strengthening effects in the composites could be classified into direct and indirect mechanisms [4] . Direct strengthening is obtained by the load being transferred from the weaker matrix to the hard reinforcements [4,7] during deformation, while indirect strengthening results from the variation of the matrix microstructures (such as Orowan strengthening mechanism, grain refinement, dislocation multiplication, etc.) in the composites by the addition of the reinforcements [8–11] . Chawla et al. [11] showed that the coefficient of thermal expansion mismatch between the SiC particles and the aluminum alloy matrix leads to formation of the dislocations along the reinforcement/matrix interfaces during cooling, and these thermal induced dislocations can effectively strengthen the matrix. It is commonly recognized that both the particle distribution and particle size have obvious effects on the mechanical properties of the composites. The voids coexisted with the clustered SiC particles and around the large sized SiC particles can be treated as pre-existed cracks [12,13] . These particles cannot transfer any load from soft matrix to the hard reinforcements, and thus, lead to the degraded mechanical properties of the composites. For a given particle volume fraction, there is a close relationship between the particle size and the deformation behavior of the composite. In general, particle size has a significant effect on the dislocation strengthening mechanism, but has almost no effect on the load transferring strengthening mechanism [6,14] . Large particles are more prone to fracture during extrusion process and tensile testing, and these fractured particles can also be treated as pre-existed cracks [15,16] , which cannot transfer any load from the matrix to the reinforcements. Lewandowski et al. [17] observed an improvement in the reinforcement distribution and better combination of the mechanical properties when the relative particle size ratio decreases from 3.3 to 1.4 in 20 vol.% SiC p reinforced Al–7Zn–2Mg–2Cu–0.14Zr (weight percentage) composites by changing the SiC particle size from 5 to 16 μm. Slipenyuk et al. [18] studied the effect of matrix-to-reinforcement particle size ratio (from 2.9 to 12.9) on the microstructures and mechanical properties of SiC p reinforced Al–6Cu–0.4Mn composites. However, their studies concentrated on the small SiC particle size (smaller than the matrix powder size). In addition, a detailed study on the effects of the process parameters on the reinforcement particle distribution fabricated by powder metallurgy technique is very important to obtain the enhanced mechanical properties. Thus, in this paper, composites containing 20 vol.% SiC particles with the size of 4.7 and 77 μm were fabricated using powder metallurgy technique with different mixing time. The purpose of this paper was to reveal the relationship between the mechanical properties of the composites and the SiC particle size and distribution. 2 Experimental The matrix alloy used in this work is an Al–3.47 wt% Cu alloy with the total impurities in the matrix being less than 0.2 wt%. The Al–Cu powders were produced by inert gas atomization, and have an average size of about 60 μm after sieving. The volume fraction of the SiC particles in the composites is 20%. The SiC reinforcements are in the form of particulate with an average diameter of about 4.7 μm and 77 μm, respectively. Al–Cu and SiC powders were ball-mixed for 2 h, 7 h, 16 h and 40 h using a V-shaped powder rotator mixer with the rotation speed of about 35 r min −1 . The balls are made by hard metal, and the mass and diameter of each ball are about 20 g and 8 mm, respectively. The ball to powder weight ratio was 1:4. The mixed powders were die-pressed at the room temperature under a pressure of 200 MPa in a cylindrical steel die with the diameter of 44 mm. The specimens were then heated in a vacuum furnace (pressure of 7 × 10 −3 Pa) to 420 °C with a heating rate of 10 °C per minute. At 420 °C, the specimens were sintered for 1 h. Then the specimens were heated to 570 °C with a heating rate of 10 °C per minute. At 570 °C, the specimens were further sintered for 5 h. After sintering, the specimens were hot extruded to rods (with a diameter of ∼14.7 mm) at 430 °C with an extrusion ratio of 9:1. The extruded rods were solution heat-treated for 6 h at 505 °C, followed by cold water quenching to the room temperature. Then the rods were artificial aging-treated for 20 h at a temperature of 177 °C. It should be noted that the selection of the heat treatment cycle was based on our previous studies (unpublished). The yield strength, tensile strength and elongation of the composites were measured by tensile testing using an Instron 8802 testing machine. At room temperature, the dog-bone-shaped tensile specimens with a gage size of 6 mm in diameter and 40 mm in length were subjected to a tensile axial load at a constant strain rate of 5 × 10 −4 s −1 . The yield stress was determined as the 0.2 pct offset. The specimen's axis is along the extrusion direction. Each point of the yield strength values has been measured on three specimens and the average value was used. The microstructures of the composites and the fracture surfaces of the tensile specimens were studied using a FEI Nano230 scanning electron microscope (SEM). 3 Results and discussion 3.1 Microstructures Figs. 1 and 2 show the SEM microstructures of the composites with two types of the particle size (4.7 and 77 μm) after mixed for different periods. It can be seen from Fig. 1 that severe SiC particle clustering can be observed in the composite with the particle size of 4.7 μm, no matter how long the mixing time is. Since the ratio of matrix/reinforcement particle sizes is comparatively large, a uniform distribution of the SiC particles in the matrix can hardly be achieved. With the increase of the mixing time, the degree of the particle clustering decreases. Note that the composite in Fig. 1 c has a large number of the pores. It is not very clear at this stage why Fig. 1 c has a large number of the pores, and a further investigation is still required in the future. Fig. 2 shows that a more uniform distribution of the SiC particles in the matrix can be obtained by decreasing the ratio of the matrix/reinforcement particle sizes. For the composite containing 77 μm sized SiC particles, only very slight particle clustering can be observed after the powders were mixed for 2 h (see Fig. 2 a) and the distribution of reinforcement is more uniform in the composite with longer mixing time as shown in Fig. 2 b–d. It is known that if the particle size is much smaller than the matrix powder size, the reinforcements can cluster very easily in the matrix and result in an inhomogeneous distribution of reinforcements. This microstructure in which the large matrix particles were surrounded by the smaller SiC particles was called “necklaced” in a previous study [19] . There is a theoretical limit for the reinforcement volume fraction for particulate reinforced composites fabricated by powder metallurgy method, and below the limit the reinforcements can distribute uniformly. Slipenyuk et al. [18] put forward a method to evaluate the maximum (critical) volume fraction of reinforcements to guarantee the best possible uniform distribution of the reinforcements in the matrix powders as follows: (1) W crit = α V SiC V Al + V SiC = α 1 − 1 + d D 3 + 2 λ + λ d D 2 + 1 λ + 2 λ d D − 1 where α = 0.18 is a parameter [18] , V SiC and V Al are the volume fractions of the SiC and Al particles, D / d is the ratio of the matrix/reinforcement particle sizes and λ is the extrusion ratio. Note that the extrusion ratio λ is 9 in this study. Table 1 shows the calculated values using Eq. (1) . It can be seen from Table 1 that when the composite is reinforced by 4.7 μm and 77 μm sized SiC powder, the critical reinforcement volume fractions are 6.0% and 17.8%, respectively. This is in agreement with the microstructural observation in Figs. 1 and 2 , in which the 77 μm sized SiC particles distribute much more uniformly in the matrix, compared to the 4.7 μm sized SiC particles. Careful observation of Fig. 2 indicates that a large number of the SiC particles fracture after extrusion, while no cracking of SiC particles in Fig. 1 can be observed after deformation (see the insets in Fig. 1 for details). During extrusion, the Al–Cu matrix deforms plastically, and the load is transferred to the hard SiC reinforcements. The hard and brittle ceramic particles will fracture once the stress concentration is beyond the fracture strength of the ceramic particles [20] . In general, larger sized particles fracture much more easily than the smaller ones during deformation [21] . The fracture of larger sized particles during extrusion has been well discussed in our previous work [22] . Therefore, numerous SiC particles are fractured with the size of 77 μm, as shown Fig. 2 , while almost no SiC particles fracture with the size of 4.7 μm, as shown in Fig. 1 . 3.2 Tensile behaviors of the composites Fig. 3 shows the mechanical properties of the composites. The yield strength, ultimate tensile strength and elongation of the composites containing small SiC particles (4.7 μm) and large SiC particles (77 μm) generally increase with increasing the mixing time. This increase is due to the decrease in the clustering degree of the SiC particles in the matrix with increasing the mixing time. With the increase of the mixing time, the SiC particles distribute more uniformly in the matrix, resulting in higher strength and ductility. However, it should be noted that the yield strength, ultimate strength and elongation of the composite containing 4.7 μm sized SiC particles with the mixing time of 16 h are unusually low. From Fig. 1 c, we know that a large number of the pores exist in the composite, and these pores substantially decrease the strength and ductility of the composite. It is not very clear at this stage why the composite containing 4.7 μm sized SiC particles after mixed for 16 h have many pores in the structures. A further investigation is still required. It can be seen that the mechanical behaviors of the composites are greatly influenced by the spatial distribution of the SiC particles. Externally applied load is generally transferred from the particle-lacking regions to the particle-rich regions in a composite, with the damage being generated in the particle-rich regions due to the higher stress concentration at small strains [23] . Since these regions quickly lose their load bearing ability, the strength of the composite with particle clustering drops to the values below that of the composites without particle clustering. In addition, the clustered particles are the sites for damage accumulation [5] and the local particle-rich regions are the most favorable nucleation sites of the cracks [1] . Clyne and Withers [1] suggested that during the growth of the crack, the particle clustering can behave in two distinct possible ways to accelerate the crack nucleation. First, they can deform collectively as a group somewhat like a single large hard particle, so that the deformation within the matrix at the center of the cluster is much less than in the composite as a whole. Secondly, the particles behave independently, such that the deformation within the cluster is much greater than in the composite overall [24] . Clustering fracture occurs at an early stage of the tensile loading process and the cracks in the fractured clusters grow to the final fracture [25] . Therefore, clustering of the reinforcement in the composite has a negative effect on the strength of the composite. In addition, when damage does occur, clustering can enhance the crack formation and propagation, and thus decrease the ductility (elongation) of the material [23] . This is in consistent with the finite element method (FEM) work carried out by Brockenbrough et al. [26] , who found that closely spaced particles resulted in high hydrostatic stresses. Fig. 3 also shows that the composites reinforced with small SiC particles (4.7 μm) exhibited significantly higher yield strength and ultimate tensile strength, while the composites reinforced with large SiC particles (77 μm) possess a higher elongation to fracture. For a composite undergoing elastic loading, a significant fraction of the stress is initially borne by the reinforcements, and the composite undergoes microplastic yielding [27] . Microplasticity in the composites takes place locally with high stress concentrations, normally along the poles of the reinforcements or at sharp corners of the reinforcements [28,29] . Generally, the particles are loaded to a higher stress with the increase in the applied strain. Compared to the matrix, the particles will usually begin to fracture prior to ultimate tensile strength of the composites because of their lower strain-to-failure. In some cases, the onset of particle fracture has occurred at stresses slightly below the yield strength of the composites [30] , and increases with increasing the size and volume fraction of the SiC particles [31] . From Fig. 2 , we know that a large number of the particle cracking have been observed after extrusion. The cracked particles in the composites cannot transfer any load since the stress is released along the cracking surfaces. Furthermore, the cracked particles will act as the microcrack initiators during deformation, leading to the decrease in the strength of the composites, as observed in Fig. 3 . Meanwhile, the nucleated voids are unable to coalesce so easily because the smaller inter-particle spacing (directly related to the particle size) imposed higher plastic constraint [32] . The formation of dislocation tangles around the particles due to plastic incompatibility between the reinforcement and matrix, and the formation of a dislocation cell structure with a cell size inversely proportional to the inter-particle spacing could also contribute to the increase in the strength with decreasing the SiC particle size [27,33] . The effect of particle size on the tensile behavior indicates an increase in the ductility with increasing the particle size. This may attribute to the increase of the inter-particle spacing with increasing the particle size, resulting in a lower plastic constraint imposed on the matrix and fewer barriers for dislocation movement and better ductility. Bhanu Prasad et al. [34] proposed an empirically equation to evaluate the relation between ultimate tensile strength of the hot pressed composite and the relative particle size ratio ( x :1): (2) σ U T S = A [ 358 − 32.5 ln ( x ) ] where A = 0.9724 is the correlation coefficient. From Fig. 3 b, we can see that the experimental determined ultimate tensile strengths of the composites with large relative particle size ratio (12:1, SiC particle size of 4.7 μm) are close to the theoretical values. The experimental determined ultimate tensile strengths approach to the theoretical values with increasing the mixing time, except for the composite with the mixing time of 16 h. The reason is that the SiC particles distribute more uniformly with increasing the mixing time. However, the experimental determined ultimate tensile strengths differ a lot from the theoretical values with small relative particle size ratio (1:1.3). According to the theoretical prediction, the composites with smaller relative particle size ratio should have higher ultimate tensile strength since the distribution of reinforcement is more uniform in the composites. However, in our study, the composites with smaller relative particle size ratio have low ultimate tensile strengths since the size of the SiC particles are too large, resulting in particle cracking during extrusion. It is widely known that the cracked particles cannot transfer any load from the matrix to the reinforcements, but act as microcrack initiators, which substantially decreases the strength and elongation. 3.3 Fracture mechanisms Figs. 4 and 5 show the SEM fracture surfaces of the composites with the SiC particles size of 4.7 μm and 77 μm, respectively. It can be seen from Fig. 4 that the fracture surfaces of composites reinforced by 4.7 μm sized SiC particles after mixed for different periods show similar characteristics, including both ductile and brittle fracture features. The fracture surfaces consist of numerous dimples on the matrix and decohesion of the SiC particles from the matrix. The dimples should be a result of the void nucleation and subsequent coalescence by strong shear deformation and fracture process on the shear plane, while the decohesion of the SiC particles can be explained by “pull-out” of the ceramic phases caused by high stress concentration. Fig. 5 shows that the SiC particles with the size of 77 μm in the composites exhibit typical brittle fracture, surrounded by the ductile dimples of the surrounding matrix. The main difference of fracture surfaces between Figs. 4 and 5 is that the “pull-out” of the SiC particles from the matrix can be clearly observed in the composites with small particles (4.7 μm), while the dominant fracture mechanism for the composites with large SiC particles (77 μm) is particle fracture. The relationship between the particle strength and particle/matrix interfacial bonding strength is the critical criterion to determine the fracture mode of the composites during deformation [35] . If the particle/matrix interfacial bonding strength is high, particle fracture usually happens during deformation. On the other hand, if the particle/matrix interfacial bonding strength is weak, decohesion between the SiC particles and the matrix will occur prior to the particle fracture. It is known that extrusion can help to decrease the number of pores and improve interfacial bonding strength of the composites. Another important feature is that extrusion results in the fracture of a large number of the SiC particles in the composites with large SiC particle size (see Fig. 2 ). The above mentioned phenomena result in the fracture mode changing from particle “pull-out” from the matrix for the composites with small SiC particle size (4.7 μm) to particle fracture for the composites with large SiC particle size (77 μm). 4 Conclusion In this paper, the combined effects of particle size and distribution on the mechanical behavior of the composites have been studied. The result indicates that small ratio between matrix/reinforcement particle sizes results in more uniform distribution of the SiC particles in the matrix. The SiC particles distribute more uniformly in the matrix with increasing the mixing time. Homogenous distribution of the SiC particles results in higher yield strength, ultimate tensile strength and ductility. Yield strength and ultimate tensile strength of the composite reinforced by 4.7 μm sized SiC particles are higher than those of composite reinforced by 77 μm sized SiC particles, while the elongation shows opposite trend with yield strength and ultimate tensile strength. The fracture surface of the composites changes from particle “pull-out” from the matrix for the composites with small SiC particle size (4.7 μm) to particle fracture for the composites with large SiC particle size (77 μm). Acknowledgements This work is supported by National Natural Science Foundation of China ( 50801068 ), Ph.D. Programs Foundation of Ministry of Education of China ( 200805331044 ) and Chinese Postdoctoral Science Foundation ( 200801345 ). References [1] T.W. Clyne P.J. Withers An Introduction to Metal Matrix Composites 1993 Cambridge University Press Cambridge [2] J.M. Torralba C.E. da Costa F. Velasco J. Mater. Proc. Technol. 133 2003 203 206 [3] B. Maruyama W.H. Hunt Jr. J. Metals 51 1999 59 61 [4] N. Chawla Y.L. Shen Adv. Eng. Mater. 3 2001 357 370 [5] S.J. Hong H.M. Kim D. Huh C. Suryanarayana B.S. Chun Mater. Sci. Eng. A 347 2003 198 204 [6] A. Slipenyuk V. Kuprin Y. Milman J.E. Spowart D.B. Miracle Mater. Sci. Eng. A 381 2004 165 170 [7] L.C. Davis J.E. Allison Metall. Trans. A 24 1993 2487 2496 [8] M. Vogelsang R.J. Arsenault R.M. Fisher Metall. Trans. A 17 1986 379 389 [9] R.J. Arsenault S.B. Wu Scripta Metall. 22 1988 767 772 [10] P.E. Krajewski J.E. Allison J.W. Jones Metall. Mater. Trans. A 24 1993 2731 2741 [11] N. Chawla C. Andres J.W. Jones J.E. Allison Metall. Mater. Trans. A 29 1998 2843 2854 [12] N. Chawla J.J. Williams R. Saha J. Light Metals 2 2002 215 227 [13] F. Tanga H. Meeks J.E. Spowart T. Gnaeupel-Herold H. Prask I.E. Anderson Mater. Sci. Eng. A 386 2004 194 204 [14] Y.W. Yan L. Geng A.B. Li Mater. Sci. Eng. A 448 2007 315 325 [15] L.C. Davis C. Andres J.E. Allison Mater. Sci. Eng. A 249 1998 40 45 [16] A. Rabiei L. Vendra T. Kishi Composites: Part A 39 2008 294 300 [17] J.J. Lewandowski C. Liu W.H. Hunt P. Kumar K. Vedula Powder Metallurgy Composites 1988 TMS Warrendale, PA 117 139 [18] A. Slipenyuk V. Kuprin Yu. Milman V. Goncharuk J. Eckert Acta Mater. 54 2006 157 166 [19] I.C. Stone P. Tsaikiropoulos A. Miravete Metal Matrix Composites, ICCM/9, Vol. 1 1993 WoodHead Publishing Ltd. pp. 271–278 [20] M.T Kiser F.W. Zok D.S. Wilkinson Acta Mater. 44 1996 3465 3476 [21] L.M. Tham M. Gupta L. Cheng Mater. Sci. Eng. A 326 2002 355 363 [22] Z. Wang M. Song C. Sun D. Xiao Y. He Mater. Sci. Eng. A 527 2010 6537 6542 [23] K.T. Conlon D.S. Wilkinson Mater. Sci. Eng. A 317 2001 108 114 [24] P.B. Prangnell S.J. Barnes S.M. Roberts P.J. Withers Mater. Sci. Eng. A 220 1996 41 56 [25] Y. Wu E.J. Lavernia Scripta Metall. 27 1992 173 178 [26] J.R. Brockenbrough S. Suresh H.A. Wienecke Acta Metall. 39 1991 735 752 [27] J.J. Williams G. Piotrowski R. Saha N. Chawla Metall. Mater. Trans. A 33 2002 3861 3869 [28] S.F. Corbin D.S. Wilkinson Acta Metall. Mater. 42 1994 1319 1327 [29] N. Chawla C. Andres J.W. Jones J.E. Allison Scripta Mater. 38 1998 1596 1600 [30] E. Marie C. Verdu G. Lormand R. Fougeres Mater. Sci. Eng. A 196 1995 135 144 [31] M. Finot Y.-L. Shen A. Needleman S. Suresh Metall. Mater. Trans. A 25 1994 2403 2420 [32] P.M. Mummery B. Derby D.J. Buttle C.B. Scruby T.W. Clyne P.J. Withers Proceedings of the 2nd European Conference on Advanced Materials, Euromat 91, Vol. 2 1991 Inst. of Metals Cambridge, UK pp. 441–447 [33] M. Manoharan J.J. Lewandowski Mater. Sci. Eng. A 150 1992 179 186 [34] V.V. Bhanu Prasad B.V.R. Bhat Y.R. Mahajan P. Ramakrishnan Mate. Sci. Eng. A 337 2002 179 186 [35] Y.-L. Shen J.J. Williams G. Piotrowski N. Chawla Y.L. Guo Acta. Mater. 49 2001 3219 3229