Author's Accepted Manuscript
Microstructures and mechanical properties of Al/Zn composites prepared by accumulative roll bonding and heat treatment C.Y. Liu, B. Zhang, P.F. Yu, R. Jing, M.Z. Ma, R.P. Liu
www.elsevier.com/locate/msea
PII: DOI: Reference:
S0921-5093(13)00568-6 http://dx.doi.org/10.1016/j.msea.2013.05.038 MSA29935
To appear in:
Materials Science & Engineering A
Received date: 18 March 2013 Revised date: 6 May 2013 Accepted date: 9 May 2013 Cite this article as: C.Y. Liu, B. Zhang, P.F. Yu, R. Jing, M.Z. Ma, R.P. Liu, Microstructures and mechanical properties of Al/Zn composites prepared by accumulative roll bonding and heat treatment, Materials Science & Engineering A, http://dx.doi.org/10.1016/j.msea.2013.05.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Microstructures and mechanical properties of Al/Zn composites prepared by accumulative roll bonding and heat treatment C.Y. Liu, B. Zhang, P.F. Yu, R. Jing, M.Z. Ma, R.P. Liu* State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao 066004, Hebei Province, China Abstract Al/Zn composites were fabricated using 1060-Al plates and Zn particles by accumulative roll bonding and heat treatment. X-ray diffraction analysis reveals that most of the Zn phase in the final Al/Zn composite disappeared in the Al matrix. Scanning electron microscopy and energy-dispersive X-ray spectroscopy analyses revealed excellent solid solution regions distribution in the final Al/Zn composite. The final Al/Zn composite showed higher hardness values of the solid solution regions than the Al matrix. Compared with ARBed monolithic 1060-Al, the Al/Zn composites had higher strength without sacrificing ductility. Keywords: Metal-matrix composites; Mechanical properties; Microstructures; Accumulative roll-bonding
*Corresponding author. address: State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao 066004, China e-mail:
[email protected] (R.P. Liu)
Tel: 0086-335-8074723 Fax: 0086-335-8074545
1
1. Introduction Aluminum-based metal matrix composites (AMMCs) are attracting considerable attention. The traditional methods used to fabricate AMMCs include powder metallurgy [1], squeeze casting [2], pressureless infiltration [3] and spray forming [4]. Lu et al. [5] and Alizadeh and Paydar [6] produced AMMCs in 2009 by accumulative roll bonding (ARB) [7], a well-known severe plastic deformation (SPD) technique. Subsequently, several AMMCs such as Al/SnO2 [8], Al/ZrO2 [8], Al/Al2O3 [8–11], Al/B4C [12,13], Al/SiC [8, 14, 15], Al/WC [16], Al/Cu [17,18], Al/W [19], Al/carbon fibers [20] and Al/intermetallic compounds [20, 21] have been fabricated by this new method because of the resulting highly uniform reinforced particle distribution and potential economic efficiency. AMMCs are characterised by low density and high tensile strength but usually have low ductility at room temperature. AMMCs with high strength and good ductility are yet to be produced by ARB. Compared with ARBed monolithic Al, ARBed AMMCs have lower ductility because of the presence of hard particles in the Al matrix. The reinforcing phases are non-plastic and always the preferential sites for crack nucleation during the tensile process. Insufficient bonding at the interface between the reinforcing phase and Al matrix results in a preferential crack propagation path that leads to plastic instability and the onset of necking. In this paper, Al/10wt.%Zn composites were fabricated by ARB and heat treatment. The composites are reinforced by Al-Zn solid solution regions. Unlike other ARBed AMMCs, the reinforcing phases in the final Al/Zn composites are also deformed during the tensile process; the preferential crack nucleation sites and crack propagation paths are eliminated; and ductility as well as strength can be achieved.
2
2. Experimental 2.1. Materials The raw materials used were as-annealed 1060-Al sheets cut parallel to the sheet-rolling direction into 150 mm × 100 mm × 1 mm pieces, as well as Zn particles less than 4 μm in size. Table 1 presents the chemical composition of 1060-Al, and Fig. 1 shows the scanning electron microscopy (SEM) image of the Zn particles used. Two pieces of the original 1060-Al sheets were degreased using acetone and then wire brushed. Subsequently, approximately 9 g of Zn particles were uniformly distributed between the two pieces of 1060-Al sheets using a scraper knife, and then the 1060-Al sheets were stacked together. The roll-bonding process was performed at room temperature without lubrication using a laboratory rolling mill 300 mm in diameter and 220 mm in barrel length. The rolling speed and rolling reduction were 0.1 m/s and 50%, respectively. Approximately 30cm3 (81 g) of Al was used in the above process. Consequently, a composite with 10 wt.% Zn particles was fabricated. To achieve uniform distribution of Zn in the Al matrix, the rolled composite was cut into two halves, and the above procedure was repeated to eight cycles without adding Zn particles (Step 1). The final sample is denoted as ARBstep1 Al/Zn. ARBstep1 Al/Zn was heated at 500 °C for 40 min in a tubular vacuum heat-treatment furnace under a vacuum atmosphere and subjected to rapid quenching in room-temperature water. The composite was subjected to three more cycles of ARB to introduce an ultrafine grained structure into the composite (Step 2). The final sample is denoted as ARBstep2 Al/Zn. 2.2. Characterisation The microstructures of the eight cycles of ARB monolithic 1060-Al (ARBed 1060-Al) and Al/Zn
3
composites were examined by SEM and transmission electron microscopy (TEM). Energy-dispersive X-ray spectroscopy (EDX) was used to characterise the chemical component of the composites. X-ray diffraction (XRD) with Cu Ka radiation was used to identify the phase composition of the composites. Hardness measurements were conducted on the rolling direction–normal direction (RD–ND) plane of the annealed 1060-Al, ARBed 1060-Al and Al/Zn composites using a Vickers microhardness (HV) tester at 10 g for 10 s. The values are the average of 50 measurement points. The tensile samples were machined from the annealed 1060-Al, ARBed 1060-Al and Al/Zn composites with the tensile axis parallel to the rolling direction. Tensile tests were performed at a strain rate of 2 × 10–4 s–1 using an Instron-5982 type test machine, and stress–strain curves were then constructed. Fracture surfaces after the tensile tests were observed by SEM to determine the failure mode. 3. Results and discussion 3.1. Structural evaluation The XRD pattern of the samples fabricated by a different process is shown in Fig. 2. Both Al and Zn are observed in the ARBstep1 Al/Zn (B). After heat treatment, the intensity of Zn peaks decreases (C). The XRD results almost do not reveal the presence of the Zn phase, which means that most Zn atoms are dissolved in the Al lattice. The Al and Zn atoms diffuse together at 500 °C, and most Zn atoms are frozen in the Al matrix during the water-quenching process. Consequently, Al-Zn solid solution regions are formed. After the second ARB process, the intensity of Zn peaks slightly increases (D). Straumal et al. [22] reported that an SPD process such as high-pressure torsion leads to the decomposition of the supersaturated solid solution in Al-Zn alloys. In this study, some Zn atoms in the Al-Zn solid solution regions may also be rejected from the Al lattice during ARB. However, the Zn phase in ARBstep2 Al/Zn is far less than that in ARBstep1 Al/Zn. Fig. 2 also shows that the Al XRD peaks are not shifted in C and
4
D. This phenomenon can be explained by the lower proportion of the Al-Zn solid solution region than that of the no-Zn atom region in the ARBstep2 Al/Zn composite. Fig. 3 shows the SEM images of the RD–ND plane of the Al/Zn composites. Unlike the microstructures of ARBed ceramic particle-reinforced AMMCs [5,6,8–16], a highly uniform reinforcing particle distribution is not observed in ARBstep1 Al/Zn [Fig. 3(a)]. The soft Zn particles are deformed with the rolling process. Some Zn particles are bonded together, and a Zn layer is produced between the two layers of Al strips after the first roll-bonding process. Subsequent ARB induces the formation of Zn fragments consisting of Zn particles. After eight ARB cycles, a composite with homogeneously distributed Zn fragments parallel to the rolling direction is achieved. Similar to the ceramic particle-reinforced AMMCs, obvious interfaces exist between the second phase and Al matrix in the ARBstep1 Al/Zn [Fig. 3(b)]. Fig. 3(c) shows that some fragments parallel to the rolling direction (white regions) also exists in ARBstep2 Al/Zn. However, the width and length are significantly larger than those of ARBstep1 Al/Zn. The colour difference in the SEM image of ARBstep2 Al/Zn is considerably weaker than that shown in Fig. 3(a). The EDX analysis results along the scanning line in Fig. 3(c) are shown in Fig. 3(d). The white and gray regions in Fig. 3(c) correspond to the Zn-rich and Zn-poor regions, respectively. The elemental distribution trend has no strict steep, and the Zn elemental content gradually decreases from the centre of the Zn-rich regions to the Zn-poor regions. Thus, no obvious interfaces exist between the two regions. Fig. 4 shows the TEM images of the RD–TD plane of annealed 1060-Al, ARBed 1060-Al and Al/Zn composites. Fig. 4(a) shows that the grain size of annealed 1060-Al is so large that a whole grain cannot be observed in the field of vision. Fig. 4(b) shows that the grain size of 1060-Al is finer and
5
becomes 0.5 μm in diameter after eight cycles of ARB. ARB introduces an ultrafine-grained structure into 1060-Al. Two phases are observed in the structure of ARBstep1 Al/Zn, i.e., Zn particles appear black and Al grains appear grey [Fig. 4(c)]. Some of the initial spherical Zn particles become finer and irregularly shaped with increased strain. Similar to the microstructures of ARBed 1060-Al, the size of Al grain in ARBstep1 Al/Zn is also refined by ARB. Fig. 4(d) shows the bright-field TEM image of the ARBstep2 Al/Zn and the selected area diffraction (SAD) patterns corresponding to points A and B. The SAD patterns reveal that the black particles are Zn. Although the composite was subjected to solid-solution treatment, some Zn particles with sizes ranging from 50 nm to 150 nm also exist in the Al matrix. Small amounts of Zn <10 nm in size are also present within the Al grains. 3.2. Mechanical properties The mechanical properties of the samples fabricated by different processes are shown in Fig. 5. Fig. 5(a) shows that the average hardness value of 1060-Al increases from 28.2 HV to 50.39 HV after eight cycles of ARB mainly because of the grain refinement and dislocation strength formed during ARB. The hardness of ARBstep1 Al/Zn is slightly higher than that of ARBed 1060-Al. The difference between the deformation capability of the Al matrix and Zn particles lead to a larger residual stress in ARBstep1 Al/Zn. More dislocations are generated in the Al matrix, which increases the hardness of ARBstep1 Al/Zn. The HV values of the ARBstep2 Al/Zn range from 53.44 HV to 89.4 HV. The hardness distribution of ARBstep2 Al/Zn corresponds to the solid-solution region distribution in the ARBstep2 Al/Zn. The values of Zn-rich regions are higher than those of the Al matrix because of solid-solution strengthening. Fig. 5(b) shows the tensile stress–strain curves of annealed 1060-Al, ARBed 1060-Al and Al/Zn
6
composites. The strengths of all samples are markedly higher than those of annealed 1060-Al. Grain refinement and strain hardening by dislocations are the two key strengthening mechanisms of ARBed 1060-Al and ARBstep1 Al/Zn. The strength of ARBstep1 Al/Zn (181 MPa) is slightly higher than that of ARBed 1060-Al (167 MPa) because of the larger residual stress of ARBstep1 Al/Zn as abovementioned. The strength of the Al/Zn composites is further enhanced after heat treatment and second ARB. The ARBstep2 Al/Zn reaches the highest tensile strength at 216 MPa. The hardness of the Al-Zn solid-solution regions is higher than that of the Al matrix. ARBstep2 Al/Zn is reinforced by the Al-Zn solid-solution regions. The strengthening mechanism of ARBstep2 Al/Zn involves the cumulative effect of the solid-solution strengthening in the Al-Zn solid-solution regions. Consequently, grain refinement and dislocation strengthening are formed during ARB. The fracture surface of ARBstep1 Al/Zn and ARBstep2 Al/Zn after tensile test is shown in Fig. 6. Dimples are found to be spread all over the secondary electron SEM (SE) images of the two composites [Figs. 6(a) and 6(c)] and clearly exhibit a typical ductile fracture in the composites. The Zn phase can be clearly distinguished from the Al matrix in the back scattered SEM (BSE) map of ARBstep1 Al/Zn, as shown in Fig. 6(b). Fig. 6(d) shows nearly no second phase in the fracture surface of ARBstep2 Al/Zn, and the reinforcing regions cannot be distinguished from the Al matrix in the BSE map. The mechanical properties of ARBed ceramic particles-AMMCs, metal particles-AMMCs, and intermetallic compounds-AMMCs have been studied by other researchers [6,9,10,11,14,17]. These AMMCs have higher tensile strength but lower tensile elongation than ARBed monolithic Al. Furthermore, the reinforcing phases of these AMMCs are non-plastic and always the preferential site for crack nucleation during the tensile process. The interface between the reinforcing phase and Al
7
matrix supplies a preferential crack propagation path that results in the lower ductility of the AMMCs. Fig. 5(b) shows that the presence of solid-solution regions in the ARBstep2 Al/Zn significantly improves the tensile strength of the Al matrix without sacrificing the tensile elongation. The reinforcing regions in this composite are also plastic, and no obvious interface exists between the reinforcing regions and Al matrix [Figs. 3(d) and 6(d)]. The preferential crack nucleation sites and crack propagation paths are eliminated during the tensile process; thus, premature failure is avoided. 4. Conclusions Al/10wt.%Zn composites were fabricated using 1060-Al plates and Zn particles by ARB and heat treatment for potential use in Al-Zn solid-solution region-reinforced AMMCs. Compared with unreinforced ARBed monolithic 1060-Al, the ARBstep2 Al/Zn composite shows high strength without sacrificing ductility. The strengthening mechanism of this composite involves the cumulative effect of grain refinement, dislocation strengthening and solid-solution strengthening. The high ductility of ARBstep2 Al/Zn can be attributed to the plasticity reinforcing regions and the absence of reinforcing region-Al matrix interfaces. Acknowledgments This work was supported by the National Basic Research Program of China (Grant no.2013CB733000) and GXB (Grant no.K0202210). References [1] M. Rahimian, N. Ehsani, N. Parvin, H.R. Baharvandi, Mater. Des. 30 (2009) 3333–3337. [2] W.G. Zhang, A.J. Song, R.P. Liu, M.Z. Ma, Mater. Sci. Eng. A 474 (2008) 225–229. [3] N.A. Travitzky, E.Y. Gutmanas, N. Claussen, Mater. Lett. 33 (1997) 47–50. [4] S. Guo, Z.L. Ning, F.Y. Cao, J.F. Sun, Trans. Nonferrous. Met. Soc. China 19 (2009) 343–348.
8
[5] C. Lu, K. Tieu, D. Wexler, J. Mater. Process. Technol. 209 (2009) 4830–4834. [6] M. Alizadeh, M.H. Paydar, J. Alloys Compd. 477 (2009) 811–816. [7] Y. Saito, N. Tsuji, H. Utsunomiya, T. Sakai, R.G. Hong, Scr. Mater. 39 (1998) 1221–1227. [8] C.W. Schmidt, C. Knieke, V. Maier, H.W. Höppel, W. Peukert, M. Göken, Scr.Mater. 64 (2011) 245–248. [9] M. Göken, H.W. Höppel, Adv. Mater. 23 (2011) 2663–2668. [10] R. Jamaati, M.R. Toroghinejad, Mater. Sci. Eng. A 527 (2010) 4146–4151. [11] M. Rezayat, A. Akbarzadeh, A. Owhadi, Composites: Part A 43 (2012) 261–267 [12] A. Yazdani, E. Salahinejad, Mater. Des. 32 (2011) 3137–3142. [13] M. Alizadeh, M.H. Paydar, Mater. Sci. Eng. A 538 (2012) 14–19. [14] M. Alizadeh, M.H. Paydar, D. Terada, N. Tsuji, Mater. Sci. Eng. A 540 (2012) 13–23. [15] M. Alizadeh, M.H. Paydar, J. Alloys Compd. 492 (2010) 231–235. [16] C.Y. Liu, Q. Wang, Y.Z. Jia, B. Zhang, R. Jing, M.Z. Ma, Q. Jing, R.P. Liu, Mater. Des. 43 (2013) 367-372. [17] C.W. Schmidt, P. Ködler, H.W. Höppel, M. Göken, Metals 1 (2011) 65–78. [18] M. Alizadeh, M. Talebian. Mater. Sci. Eng. A 558 (2012) 331-337. [19] C.Y. Liu, Q. Wang, Y.Z. Jia, B. Zhang, R. Jing, M.Z. Ma, Q. Jing, R.P. Liu, Mater. Sci. Eng. A 547 (2012) 120–124. [20] T. Hausöl, V. Maier, C.W. Schmidt, M. Winkler, H.W. Höppel, M. Göken. Adv. Eng. Mater. 12 (2010) 740–746. [21] C.Y. Liu, R. Jing, Q. Wang, B. Zhang, Y.Z. Jia, M.Z. Ma, R.P. Liu, Mater. Sci. Eng. A 558 (2012) 510-516.
9
[22] B. Straumal, R. Valiev, O. Kogtenkova, P. Zieba, T. Czeppe, E. Bielanska, M. Faryna, Acta Mater. 56 (2008) 6123-6131.
Figure and Table captions Fig. 1. SEM image of the Zn particles used Fig. 2. XRD patterns of (A) ARBed 1060-Al, (B) ARBstep1 Al/Zn, (C) ARBstep1 Al/Zn subjected to heat treatment and (D) ARBstep2 Al/Zn Fig. 3. SEM images of ARBstep1 Al/Zn: (a) low magnification and (b) high magnification. (c) SEM image of ARBstep2 Al/Zn. (d) EDX analyses along the scanning line in (c) Fig. 4. TEM images of (a) annealed 1060-Al, (b) ARBed 1060-Al, (c) ARBstep1 Al/Zn and (d) ARBstep2 Al/Zn Fig. 5. (a) Vickers microhardness and (b) stress–strain curves of the annealed 1060-Al, ARBed 1060-Al and Al/Zn composites Fig. 6. (a) SE and (b) BSE map obtained from the fracture surfaces of ARBstep1 Al/Zn. (c) SE and (d) BSE map obtained from the fracture surfaces of ARBstep2 Al/Zn after tensile test
Table 1. Chemical composition of the Al sheets used
10
Fig.1
11
Fig.2
12
Fig.3
13
Fig.4
14
Fig.5
15
Fig. 6
Table 1. Element
Fe
Mg
Si
Zn
Ti
Cu
Al
Wt (%)
0.03
0.03
0.25
0.05
0.03
0.05
99.56
16