- Email: [email protected]
Influence of equal-channel angular pressing on aging precipitation in 7050 Al alloy
Intermetallics 55 (2014) 49e55
Contents lists available at ScienceDirect
Intermetallics journal homepage: www.elsevier.com/locate/intermet
Influence of equal-channel angular pressing on aging precipitation in 7050 Al alloy M.H. Li a, *, Y.Q. Yang a, Z.Q. Feng a, G.H. Feng a, B. Huang a, Y.X. Chen a, M. Han a, J.G. Ru b a b
State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an 710072, PR China Beijing Institute of Aeronautical Materials, Beijing 100095, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 December 2013 Received in revised form 8 June 2014 Accepted 9 July 2014 Available online 30 July 2014
7050 Al alloy was successfully processed by equal-channel angular pressing (ECAP) at room temperature (RT). The effect of ECAP on the subsequent aging precipitation behavior was investigated by using transmission electron microscopy (TEM) and high resolution transmission electron microscopy (HRTEM). The results reveal that the kinds, spatial distribution and sizes of precipitates in the unECAPed and the ECAPed samples are different. ECAP accelerates the process of aging precipitation and results in the broadening of precipitate size distribution. ECAP can produce not only deformation heat but also internal defects such as excess vacancies and high density of dislocations when the sample passes through the main deformation zone. Deformation heat can lead to pre-precipitation, forming a small amount of GPII zones during ECAP processing. Strain-induced excess vacancies make solute segregation along dislocations by the mechanism of nonequilibrium segregation. High density dislocations mainly accelerate the process of aging precipitation. Besides, dislocations also induce the competition between homogeneous precipitation and heterogeneous precipitation on dislocations due to the flow of solutes and vacancies towards dislocations. © 2014 Elsevier Ltd. All rights reserved.
Keywords: B. Precipitates B. Strain-aging C. Heat treatment F. Electron microscopy, transmission
1. Introduction Severe plastic deformation (SPD) is an effective method of producing metals and alloys with ultrafine-grained and/or nanocrystalline structures [1,2]. Among various SPD techniques currently available, ECAP is one of the most attractive methods to impose extremely large plastic strain to bulk materials [1,2]. It is well known that the strengthening mechanism by ECAP is grain refinement strengthening and strain hardening. Moreover, previous studies [3,4] have shown that only one pass of ECAP is enough to improve the strength of Al alloys. More recently, there has been a growing interest in post-ECAP aging treatment for age-hardenable Al alloys to achieve a combination of strengthening from grain refinement, strain hardening and precipitation hardening [3e7]. Some researchers [8e11] have referred to the influences of plastic deformation on precipitation in age-hardenable Al alloys. Plastic deformation not only promotes the aging precipitation processes [8e10], but also forms the dynamic precipitation of GP zones during deformation [10,11]. The accelerated kinetics is mainly concerned with excess vacancies and a high density of dislocations which are generated during deformation. On the one
* Corresponding author. Tel./fax: þ86 29 88460499. E-mail address: [email protected] (M.H. Li). http://dx.doi.org/10.1016/j.intermet.2014.07.005 0966-9795/© 2014 Elsevier Ltd. All rights reserved.
hand, excess vacancies increases solute diffusion since there exist solute-vacancy binding energies in Al alloys [12]. On the other hand, dislocations act as nucleation sites for precipitates and provide pipe-diffusion paths [8,10]. Although the influences of plastic deformation on precipitation in age-hardenable Al alloys have been investigated [8e11], besides the influence of the vacancy and dislocation on precipitation, there may also exist other mechanisms, such as the influence of temperature rise on precipitation. Moreover, the effect of plastic deformation on aging precipitation is extremely interrelated with experimental processes (such as deformation parameters and heat treatment conditions). In the paper, therefore, we study the influence of SPD on aging precipitation in 7050 Al alloy by experimental process which is different from previous studies. A 7050 Al alloy sample was subjected to ECAP immediately after solution treatment, and then artificial aging at 160 C for 6 h. The kinds, spatial distribution and sizes of precipitates were analyzed by using TEM and HRTEM. In addition, the mechanism of the effect of ECAP on aging precipitation was discussed in detail.
2. Experimental The 7050 Al alloy was chosen for investigation and the nominal chemical composition of the alloy is (wt%):
50
M.H. Li et al. / Intermetallics 55 (2014) 49e55
TEM operated at 300 kV. For overviewing the grain size and grain size distribution of the alloy, electron back-scattered diffraction (EBSD) was conducted by using HKL channel 5 equipped in a Zeiss Supra 55 scanning electron microscope (SEM). The precipitate size distributions of the unECAPed and the ECAPed samples aged at 160 C for 6 h were obtained by using VNT QuantLab-MG software. 3. Results 3.1. Microstructures after solution treatment
Fig. 1. Schematic illustration of the ECAP die. The main deforming zone is indicated in the black area.
Table 1 Details of ECAP and heat treatment processes used in current study. Sample
Processing
ECAPed
Solution treatment (475 C/1 h) þ ECAP (one pass at RT) þ aging treatment (160 C/6 h) Solution treatment (475 C/1 h) þ aging treatment (160 C/6 h)
unECAPed
Ale6.29Zne2.22Mge2.28Cue0.15Zr, with a small amount of Fe, Si, Mn, Ti and Cr. The cast ingot was homogenized at 460 C for 24 h, and then hot rolled to a 30 mm thin plate, following by cutting into short billets with the dimension of 20 mm 20 mm 60 mm. These billets were solution treated at 475 C for 1 h and subsequently water quenched. The ECAP processing was conducted at RT immediately after water quenching through a die with a plunger speed of 6 mm/s. The ECAP die has the angle of intersection of the two channels F ¼ 90 and the angle subtended by the arc of curvature at the point of intersection J ¼ 20 , as shown in Fig. 1. The billet and the die were well lubricated using MoS2-containing grease. The number of ECAP pass is one. For comparison purposes, both the unECAPed and the ECAPed samples were aged at 160 C for 6 h. The full details of these two processes are summarized in Table 1. All TEM samples were subsequently prepared by mechanical grinding and punching into 3 mm disks in diameter. The disks were finally thinned using twin jet electropolishing with an electrolyte of 30% nitric acid and 70% methanol at 15 V at 25 C. Microstructure characterization was carried out by Tecnai F30 G2
Fig. 2 presents an EBSD band contrast map and a bright-field (BF) TEM image showing the typical microstructure of hot-rolled 7050 Al alloy which was solution treated at 475 C for 1 h and then water quenched to RT. One can see from the EBSD band contrast map in Fig. 2(a) that, although there exist elongated grains of 2e8 mm in width and 10e24 mm in length due to hot rolling, a lot of equiaxial grains of about 2e5 mm have been formed after the solution treatment. Fig. 2(b) shows grain size distribution histogram from Fig. 2(a). The grain equivalent diameter ranges from 2 mm to 17 mm. Fig. 2(c) shows some equiaxial grains of about 4 mm in diameter observed by TEM, and the grains have low dislocation density. 3.2. Microstructures after ECAP Fig. 3 presents the TEM micrographs with low and high magnification, revealing the microstructure after ECAP at RT for one pass. The subgrains were apparently elongated along the shear direction. The size of the elongated grains ranges from 100 to 500 nm in the longitudinal direction, and is about 100 nm in width, as shown in Fig. 3(a). High density of dislocations was homogeneously generated in the ECAPed alloy, as shown in Fig. 3(b). Fig. 3(c) is a typical HRTEM image along <011>Al direction showing the morphological and crystallographic features of the GPII zones [13]. Some planar precipitates are fully coherent with the Al matrix, parallel to {111}Al planes, with the thickness of about 1e3 times of {111}Al atomic plane spacing and the length of about 4e7 nm. Additionally, weak diffraction streaks are observed along {111}Al in the Fast Fourier transformation (FFT) spectrum of Fig. 3(c), indicating the GPII zones formed during ECAP processing. 3.3. Microstructures of the samples after aging treatment Fig. 4 shows the characterization of the precipitate microstructure of the unECAPed and the ECAPed samples aged at 160 C for 6 h Fig. 4(a) and (b) present BF TEM images of the two samples. In the unECAPed sample, fine dispersed disk-shaped precipitates with
Fig. 2. EBSD band contrast map (a), grain size distribution histogram (b) and BF TEM micrograph (c), showing the grains of the 7050 Al alloy after solid solution treatment.
M.H. Li et al. / Intermetallics 55 (2014) 49e55
51
Fig. 3. TEM micrographs of 7050 Al alloy after ECAP at RT for one pass showing the elongated grains (a), high density of dislocations (b), and some GPII zones in Al matrix from HRTEM micrograph of [011]Al projection (c). The inset is the corresponding FFT spectrum.
the size of about 2e6 nm in diameter are homogeneously distributed (Fig. 4(a)). To observe clearly the precipitates in the ECAPed sample, the contrast of dislocations are avoided by tilting the sample holder in Fig. 4(b). In the ECAPed sample, the large lathshaped precipitates have length of about 15e33 nm and thickness of about 7e16 nm, and the small disk-shaped precipitates have the size of about 3e10 nm in diameter. Fig. 4(c) and (d) shows typical <112>Al HRTEM images taken from the two samples. Fig. 4(c) shows some fine h0 precipitates (marked by a white arrow) in the unECAPed sample. The existence of the h0 phase is confirmed by the weak diffraction streaks at 1/3 and 2/3 {220}Al position in the FFT spectrum. These precipitates are about 5e8 nm long and about 2e4 nm thick. However, in the ECAPed sample, HRTEM image indicates that two types of precipitates coexist, namely h0 (marked by a white arrow) and h (marked by a black arrow) (Fig. 4(d)). Separate diffuse streaks and diffraction spots at 1/3 and 2/3 {220}Al position can be observed in the FFT spectrum, indicating the present of h0 and h phases. In this situation, the h0 precipitates have a length of about 4e7 nm and a thickness of about 2e3 nm, and h precipitates are about 15e20 nm long and about 4e7 nm thick. Fig. 4(e) and (f) shows the precipitate size histograms from Fig. 4(a) and (b) of TEM images, respectively. Two important phenomena can be observed from Fig. 4(e) and (f). First, the average precipitate equivalent diameter in the ECAPed sample is larger than that in the unECAPed sample after the same aging treatment (i.e., for 7.1 nm vs 4.0 nm). Second, the precipitate size distribution
widens in the ECAPed sample. In the unECAPed sample, the precipitate equivalent diameter ranges from 1 nm to 7 nm. However, in the ECAPed sample, the precipitate equivalent diameter ranges from 3 nm to 23 nm. Fig. 5(a) and (b) shows <011>Al selected area electron diffractions (SAEDs) patterns taken from the unECAPed and the ECAPed samples aged at 160 C for 6 h, respectively. Weak diffraction spot at the 1/3 and 2/3 {220}Al position (marked by a white arrow) indicates that h0 precipitates have formed in the unECAPed sample (Fig. 5(a)). However, two separate diffraction spots appears respectively at 1/3 and 2/3 {022}Al position (marked by a white arrow and a black arrow) in Fig. 5(b), indicating that h0 and h precipitates coexist in the aged ECAPed sample. 4. Discussion 4.1. Influence of temperature rise on precipitation According to Refs. [14e17], plastic deformation can lead to the temperature rise of the workpiece because of the heat generated by mechanical work and the friction between the ECAP die and the workpiece. Kim [16] proposed an equation for the temperature rise DT of the workpiece during ECAP processing:
DT ¼
.pffiffiffi 0:9s3 þ 0:5m s 3 uðA=VÞDt rC þ ðA=VÞhDt
(1)
52
M.H. Li et al. / Intermetallics 55 (2014) 49e55
Fig. 4. Characterization of the precipitate microstructure of the unECAPed sample (a, c, e) and the ECAPed sample (b, d, f) aged at 160 C for 6 h, showing BF TEM micrographs (a, b), HRTEM micrographs (c, d) and precipitate size histograms (e, f). The insets are the corresponding FFT spectrums.
where DT is the temperature increment of the workpiece in the main deforming zone, s is the stress, 3 is the strain imposed after pressing, m is the friction factor, u is the relative velocity between the die and the workpiece, h is the heat transfer coefficient between the workpiece and the die, A is the outer surface area of the domain contacting the die, V is the volume of the main deforming zone, Dt is the dwell time of the domain within the deforming zone, r is the density of the workpiece, and C is its heat capacity, respectively. When a sample is pressed through the die during ECAP processing, a strain, 3 , is introduced [18]:
3
¼N
2 cotðf=2 þ j=2Þ þ jcosecðf=2 þ j=2Þ pffiffiffi 3
(2)
where 3 is the strain imposed after multiple pressing, N is the number of pressing passes. For the present experiment, the sample was pressed using a die of F ¼ 90 , J ¼ 20 for one pass to give a strain 3 ¼ 1.0546. The average dwell time (Dt) of the domain within the deforming zone is:
M.H. Li et al. / Intermetallics 55 (2014) 49e55
53
Fig. 5. SAED patterns for samples aged at 160 C for 6 h: (a) the unECAPed sample and (b) the ECAPed sample. The inset is magnification of the rectangular region in (b).
Dt ¼
d j pffiffiffi u 2p
(3)
where d is the diameter of the cylindrical workpiece. For the present experiment, the cuboid workpiece with the size of 20 mm 20 mm 60 mm can be approximated as the cylindrical workpiece with a diameter 22.57 mm and 60 mm in length. u ¼ 6 mm/s. So Dt is evaluated to be 0.3 s. The parameter values for 7050 Al alloy, s ¼ 400 MPa, 3 ¼ 1.0546, r ¼ 2830 kg/m3, C ¼ 860 J/Kg K, m ¼ 0.2, and h ¼ 2000 N/m s K are used for calculation. By substituting these values into equation (1), DT ¼ 151 C can be finally obtained. Based on Refs. [14e17], there are a good agreement between the calculated, the measured and the simulated temperature rise of Al and Al alloys during ECAP processing. So, in the study, it is credible that the temperature is increased abruptly to 151 C in the 7050 Al alloy immediately upon passing through the shearing plane. Moreover, as Al alloy is a good thermal conductor, the temperature of the workpiece increases rapidly from the initial 25 C to a peak value of about 151 C, then decreases to 25 C in a few minutes during ECAP processing. Since the temperature rise may reach 151 C in 7050 Al alloy when the sample passed through the main deforming zone during ECAP processing, some GPII zones form due to the process of dynamic aging, as shown in Fig. 3(c). During post-ECAP artificial aging, the existing GPII zones can accelerate formation of metastable phase h0 . So, these GPII zones contribute to larger average precipitates size and the variation in precipitate size distribution in the post-ECAP aged sample. 4.2. Influence of vacancies on precipitation Solute segregation occurs along dislocations due to straininduced excess vacancies through the mechanism of nonequilibrium segregation [19,20]. The three necessary conditions for nonequilibrium segregation are 1) the existence of excess vacancies, 2) the presence of vacancy sinks, and 3) a positive vacancyimpurity binding energy. These three conditions are fully satisfied in the 7050 Al alloy subjected to ECAP, as discussed subsequently. 1) The existence of excess vacancies. SPD can create excess vacancies [21e23]. Deformation dependence of the vacancy concentration c of point defects may be written as [24]:
dc s ¼ d3 3m
(4)
where 3 is the strains, m is the shear modulus, s is the stress value and c is the vacancy concentration. This law is obeyed fairly well in face centered cubic metals [24]. Ref. [25] had supposed that the vacancy concentration is about 104 for strains 3 z 1, which is comparable to the equilibrium value at the melting point. 2) The presence of vacancy sinks. It is well known that dislocations are perfect vacancy sinks. ECAP can introduce high density of dislocations, as shown in Fig. 3(b). 3) A positive vacancy-impurity binding energy, Eb. The values of Eb can be calculated by using elasticity theory. According to the study of Cottrell [26], the following formula approximately describes the binding energy Eb of a vacancy with a foreign atom:
Eb ¼ 8pmr03 l2 ;
l ¼ ±ðr1 r0 Þ=r0
(5)
where l is the misfit of the foreign atom with matrix lattice, m is the shear modulus of the matrix, r0 is the matrix atom radius, and r1 is the foreign atom radius. Therefore, the binding energy of a vacancy with solute atom in 7050 Al alloy can be calculated as Mg Eb ¼ 0.28 eV, EbCu ¼ 0.22 eV and EbZn ¼ 0.06 eV. Based on the theory of nonequilibrium segregation, solute atoms gradually segregate to dislocations. Since dislocations serve as an ideal sink of vacancy, the level of excess vacancies at dislocation is zero. So, soluteevacancy complex gradients could develop near dislocations. Subsequently, soluteevacancy complexes diffuse towards dislocations due to the concentration gradients, and then the soluteevacancy complexes are decomposed into vacancies and solute atoms (Mg, Cu and Zn). Meanwhile, the vacancies will annihilate at dislocations leaving behind the solute atoms. As the increase of solute concentration at dislocations, a subsequent desegregation occurs. However, desegregation of solute is negligible because the diffusivity of single solute is much slower than that of soluteevacancy complex [19,27,28]. As a result, solute atoms gradually segregate to dislocations. 4.3. Influence of dislocations on precipitation ECAP introduces high density of dislocations both at the grain/ subgrain boundaries, as seen in Fig. 3(b). For Al alloy materials subjected to ECAP processing, dislocations density r is up to
54
M.H. Li et al. / Intermetallics 55 (2014) 49e55
1013~1014/m2 after one pass ECAP [5]. The increased dislocation density has two aspects to influence the precipitation reaction of 7050 A1 alloy. The presence of dislocations induces not only faster and coarser precipitation on dislocations but also a large precipitate size gradient. In the ECAPed sample, precipitation was strongly accelerated in aging process, as shown in Fig. 4(b). The average size of the precipitates in the ECAPed sample (Fig. 4(f)) is much larger than that in the unECAPed sample (Fig. 4(e)). According to the HRTEM images and the diffraction observed near 1/3 and 2/3 {220} in the SAEDs shown in Fig. 4(c, d) and 5, the metastable phases h0 and the equilibrium phases h coexist in the ECAPed sample whereas only the metastable phases h0 are present in the unECAPed sample aged at 160 C for 6 h. Moreover, the precipitate size distribution becomes wide in the ECAPed sample (see Fig. 4e, f). In the unECAPed sample, the precipitate equivalent diameter ranges from 1 nm to 7 nm. However, in the ECAPed sample, the precipitate equivalent diameter ranges from 3 to 23 nm. The two phenomena indicate that dislocations not only lead to nucleation and growth of larger precipitates on dislocations, but also modify homogeneous precipitation in the matrix surrounding dislocations. This can be explained from the following two aspects.
treatment is a complex precipitation process, since ECAP generates deformation heat, large quantities of vacancies and high density of dislocations. Some conclusions are summarized as follows:
1) Dislocations can reduce the activation energy for precipitation by reducing the strain energy or interface energy. Thus, dislocations act readily as heterogeneous nucleation sites [8]. Moreover, dislocations increase diffusivities of precipitateforming elements due to short circuit diffusion along dislocations [29], resulting in the rapid growth and coarsening of precipitates on dislocations.
Acknowledgments
At the early stages of artificial aging, the vacancyesolute complexes gradually migrated to dislocations due to nonequilibrium segregation. In the vicinity of dislocations, frequent collisions between these soluteevacancy complexes released vacancies which finally annihilated at dislocations leaving behind mobile complexes of solute atoms which may or may not also include vacancies. As a result, fine soluteerich clusters are formed at the beginning of artificial aging which act as GPII zones nuclei on subsequent aging. Subsequent evolution of the microstructure involves the replacement of the GPII zones with more stable phases. 2) The influence of dislocation on precipitation can't just be considered from the heterogeneous precipitation on dislocations, as described in Section 4.2. On the one hand, large quantities of solutes flow to dislocations by the mechanism of nonequilibrium segregation [19,20]. The solute concentration gradient forms around the dislocations. On the other hand, dislocations are also ideal vacancy sinks [19,20,30]. Therefore, there exists the vacancy concentration gradient around the dislocations. Since the concentration and distribution of vacancy have the strong influence on the homogeneous nucleation of precipitates in AleZneMgeCu alloys [31,32], the presence of high density of dislocations will influence homogeneous precipitation around them. As a result, the flow of solutes and vacancies to dislocations leads to competition between homogeneous precipitation in matrix and heterogeneous precipitation on dislocations, and subsequently a large precipitate size gradient has been formed. 5. Conclusions The influence of ECAP at RT on aging precipitation of 7050 Al alloy was investigated by using TEM and HRTEM. Post-ECAP aging
1) Due to deformation heat induced by SPD, the temperature rise may reach 151 C when the sample passes through the main deforming zone during ECAP processing. Some GPII zones can be observed after ECAP, which will accelerate precipitation during subsequent artificial aging, and induce the variation in precipitate size distribution. 2) High density of dislocations can accelerate nucleation as well as growth of precipitate phases. This leads to coarser precipitation on dislocations in the process of post-ECAP aging in 7050 Al alloy. 3) Although the effect of ECAP on aging precipitation is mainly heterogeneous precipitation on dislocations, due to the flow of solutes and vacancies to dislocations by the mechanism of nonequilibrium segregation, there exists competition between homogeneous precipitation in matrix and heterogeneous precipitation on dislocations. As a result, the precipitate size distribution widens.
The authors would like to acknowledge the financial support of the 111 Project (B08040) of China and the Natural Science Foundation of China (51071125, 51271147 and 51201135).
References [1] Valiev RZ, Langdon TG. Principles of equal-channel angular pressing as a processing tool for grain refinement. Prog Mater Sci 2006;51:881e981. [2] Valiev RZ, Islamgaliev RK, Alexandrov IV. Bulk nanostructured materials from severe plastic deformation. Prog Mater Sci 2000;45:103e89. [3] Kim WJ, Kim JK, Kim HK, Park JW, Jeong YH. Effect of post equal-channelangular-pressing aging on the modified 7075 Al alloy containing Sc. J Alloys Compd 2008;450:222e8. [4] Kim WJ, Chung CS, Ma DS, Hong SI, Kim HK. Optimization of strength and ductility of 2024 Al by equal channel angular pressing (ECAP) and post-ECAP aging. Scr Mater 2003;49:333e8. [5] Zhao YH, Liao XZ, Jin Z, Valiev RZ, Zhu YT. Microstructures and mechanical properties of ultrafine grained 7075 Al alloy processed by ECAP and their evolutions during annealing. Acta Mater 2004;52: 4589e99. [6] Zhao YH, Liao XZ, Cheng S, Ma E, Zhu YT. Simultaneously increasing the ductility and strength of nanostructured alloys. Adv Mater 2006;18: 2280e3. [7] Horita Z, Ohashi K, Fujita T, Kaneko K, Langdon TG. Achieving high strength and high ductility in precipitation-hardened alloys. Adv Mater 2005;17: 1599e602. [8] Gubicza J, Schiller I, Chinh NQ, Illy J, Horita Z, Langdon TG. The effect of severe plastic deformation on precipitation in supersaturated AleZneMg alloys. Mater Sci Eng A 2007;460e461:77e85. chet Y, Chemin JL, Hutchinson CR. In situ eval[9] Deschamps A, Fribourg G, Bre uation of dynamic precipitation during plastic straining of an AleZneMgeCu alloy. Acta Mater 2012;60:1905e16. [10] Deschamps A, De Geuser F, Horita Z, Lee S, Renou G. Precipitation kinetics in a severely plastically deformed 7075 aluminium alloy. Acta Mater 2014;66: 105e17. [11] Han WZ, Chen Y, Vinogradov A, Hutchinson CR. Dynamic precipitation during cyclic deformation of an underaged AleCu alloy. Mater Sci Eng A 2011;528: 7410e6. [12] Wolverton C. Soluteevacancy binding in aluminum. Acta Mater 2007;55: 5867e7872. [13] Berg LK, Gjønnes J, Hansen V, Li XZ, Knutson-Wedel M, Waterloo G, et al. GPzones in AleZneMg alloys and their role in artificial aging. Acta Mater 2001;49:3443e51. [14] Yamaguchi D, Horita Z, Nemoto M, Langdon TG. Significance of adiabatic heating in equal-channel angular pressing. Scr Mater 1999;41: 791e6. [15] Pei QX, Hu BH, Lu C, Wang YY. A finite element study of the temperature rise during equal channel angular pressing. Scr Mater 2003;49:303e8. [16] Kim HS. Prediction of temperature rise in equal channel angular pressing. Mater Trans 2001;42:536e8.
M.H. Li et al. / Intermetallics 55 (2014) 49e55 [17] Hosford WF, Caddell RM. Metal forming: mechanics and metallurgy. 3rd ed. New York: Cambridge University Press; 2007. [18] Iwahashi Y, Wang J, Horita Z, Nemoto M, Langdon TG. Principle of equalchannel angular pressing for the processing of ultra-fine grained materials. Scr Mater 1996;35:143e6. [19] Liu WJ. A new theory and kinetic modeling of precipitation of Nb(CN) in microalloyed strain-induced austenite. Metall Mater Trans A 1995;26: 1641e57. [20] Liu WC, Chen ZL, Yao M. Effect of cold rolling on the precipitation behavior of d phase in inconel 718. Metall Mater Trans A 1999;30:31e40. [21] Straumal BB, Baretzky B, Mazilkin AA, Phillipp F, Kogtenkova OA, Volkovb MN, et al. Formation of nanograined structure and decomposition of supersaturated solid solution during high pressure torsion of AleZn and AleMg alloys. Acta Mater 2004;52:4469e78. [22] Sauvage X, Wetscher F, Pareige P. Mechanical alloying of Cu and Fe induced by severe plastic deformation of a CueFe composite. Acta Mater 2005;53: 2127e35. [23] Su LH, Lu C, He LZ, Zhang LC, Guagliardo P, Tieu AK, et al. Study of vacancytype defects by positron annihilation in ultrafine-grained aluminum severely deformed at room and cryogenic temperatures. Acta Mater 2012;60: 4218e28.
55
[24] Saada G. Interaction of dislocations and hardening production of point defects in fcc metals. Acta Metall 1961;9:166e8. [25] Friedel J. Dislocations. Oxford: Pergamon Press; 1964. [26] Cottrell AH. An introduction to metallurgy. London: Edward Arnold; 1967. [27] Xu TD, Song SH, Shi HZ, Gust W, Yuan ZX. A method of determining the diffusion-coefficient of vacancy solute atom complexes during the segregation to grain-boundaries. Acta Metall Mater 1991;39:3119e24. [28] Williams TM, Stoneham AM, Harries DR. The segregation of boron to grain boundaries in solution-treated Type 316 austenitic stainless steel. Met Sci 1976;10:14e9. [29] Hoyt JJ. On the coarsening of precipitates located on grain boundaries and dislocations. Acta Metall 1991;39:2091e8. chet Y. Influence of predeformation on ageing in an [30] Deschamps A, Livet F, Bre AleZneMg alloydI. Microstructure evolution and mechanical properties. Acta Mater 1999;47:281e92. [31] Girifalco LA, Herman H. A model for the growth of GuinierePreston zonesdthe vacancy pump. Acta Metal 1965;13:583e90. [32] Embury JD, Nicholson RB. The nucleation of precipitates: the system AleZneMg. Acta Metall 1965;13:403e17.
Contents lists available at ScienceDirect
Intermetallics journal homepage: www.elsevier.com/locate/intermet
Influence of equal-channel angular pressing on aging precipitation in 7050 Al alloy M.H. Li a, *, Y.Q. Yang a, Z.Q. Feng a, G.H. Feng a, B. Huang a, Y.X. Chen a, M. Han a, J.G. Ru b a b
State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi'an 710072, PR China Beijing Institute of Aeronautical Materials, Beijing 100095, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 December 2013 Received in revised form 8 June 2014 Accepted 9 July 2014 Available online 30 July 2014
7050 Al alloy was successfully processed by equal-channel angular pressing (ECAP) at room temperature (RT). The effect of ECAP on the subsequent aging precipitation behavior was investigated by using transmission electron microscopy (TEM) and high resolution transmission electron microscopy (HRTEM). The results reveal that the kinds, spatial distribution and sizes of precipitates in the unECAPed and the ECAPed samples are different. ECAP accelerates the process of aging precipitation and results in the broadening of precipitate size distribution. ECAP can produce not only deformation heat but also internal defects such as excess vacancies and high density of dislocations when the sample passes through the main deformation zone. Deformation heat can lead to pre-precipitation, forming a small amount of GPII zones during ECAP processing. Strain-induced excess vacancies make solute segregation along dislocations by the mechanism of nonequilibrium segregation. High density dislocations mainly accelerate the process of aging precipitation. Besides, dislocations also induce the competition between homogeneous precipitation and heterogeneous precipitation on dislocations due to the flow of solutes and vacancies towards dislocations. © 2014 Elsevier Ltd. All rights reserved.
Keywords: B. Precipitates B. Strain-aging C. Heat treatment F. Electron microscopy, transmission
1. Introduction Severe plastic deformation (SPD) is an effective method of producing metals and alloys with ultrafine-grained and/or nanocrystalline structures [1,2]. Among various SPD techniques currently available, ECAP is one of the most attractive methods to impose extremely large plastic strain to bulk materials [1,2]. It is well known that the strengthening mechanism by ECAP is grain refinement strengthening and strain hardening. Moreover, previous studies [3,4] have shown that only one pass of ECAP is enough to improve the strength of Al alloys. More recently, there has been a growing interest in post-ECAP aging treatment for age-hardenable Al alloys to achieve a combination of strengthening from grain refinement, strain hardening and precipitation hardening [3e7]. Some researchers [8e11] have referred to the influences of plastic deformation on precipitation in age-hardenable Al alloys. Plastic deformation not only promotes the aging precipitation processes [8e10], but also forms the dynamic precipitation of GP zones during deformation [10,11]. The accelerated kinetics is mainly concerned with excess vacancies and a high density of dislocations which are generated during deformation. On the one
* Corresponding author. Tel./fax: þ86 29 88460499. E-mail address: [email protected] (M.H. Li). http://dx.doi.org/10.1016/j.intermet.2014.07.005 0966-9795/© 2014 Elsevier Ltd. All rights reserved.
hand, excess vacancies increases solute diffusion since there exist solute-vacancy binding energies in Al alloys [12]. On the other hand, dislocations act as nucleation sites for precipitates and provide pipe-diffusion paths [8,10]. Although the influences of plastic deformation on precipitation in age-hardenable Al alloys have been investigated [8e11], besides the influence of the vacancy and dislocation on precipitation, there may also exist other mechanisms, such as the influence of temperature rise on precipitation. Moreover, the effect of plastic deformation on aging precipitation is extremely interrelated with experimental processes (such as deformation parameters and heat treatment conditions). In the paper, therefore, we study the influence of SPD on aging precipitation in 7050 Al alloy by experimental process which is different from previous studies. A 7050 Al alloy sample was subjected to ECAP immediately after solution treatment, and then artificial aging at 160 C for 6 h. The kinds, spatial distribution and sizes of precipitates were analyzed by using TEM and HRTEM. In addition, the mechanism of the effect of ECAP on aging precipitation was discussed in detail.
2. Experimental The 7050 Al alloy was chosen for investigation and the nominal chemical composition of the alloy is (wt%):
50
M.H. Li et al. / Intermetallics 55 (2014) 49e55
TEM operated at 300 kV. For overviewing the grain size and grain size distribution of the alloy, electron back-scattered diffraction (EBSD) was conducted by using HKL channel 5 equipped in a Zeiss Supra 55 scanning electron microscope (SEM). The precipitate size distributions of the unECAPed and the ECAPed samples aged at 160 C for 6 h were obtained by using VNT QuantLab-MG software. 3. Results 3.1. Microstructures after solution treatment
Fig. 1. Schematic illustration of the ECAP die. The main deforming zone is indicated in the black area.
Table 1 Details of ECAP and heat treatment processes used in current study. Sample
Processing
ECAPed
Solution treatment (475 C/1 h) þ ECAP (one pass at RT) þ aging treatment (160 C/6 h) Solution treatment (475 C/1 h) þ aging treatment (160 C/6 h)
unECAPed
Ale6.29Zne2.22Mge2.28Cue0.15Zr, with a small amount of Fe, Si, Mn, Ti and Cr. The cast ingot was homogenized at 460 C for 24 h, and then hot rolled to a 30 mm thin plate, following by cutting into short billets with the dimension of 20 mm 20 mm 60 mm. These billets were solution treated at 475 C for 1 h and subsequently water quenched. The ECAP processing was conducted at RT immediately after water quenching through a die with a plunger speed of 6 mm/s. The ECAP die has the angle of intersection of the two channels F ¼ 90 and the angle subtended by the arc of curvature at the point of intersection J ¼ 20 , as shown in Fig. 1. The billet and the die were well lubricated using MoS2-containing grease. The number of ECAP pass is one. For comparison purposes, both the unECAPed and the ECAPed samples were aged at 160 C for 6 h. The full details of these two processes are summarized in Table 1. All TEM samples were subsequently prepared by mechanical grinding and punching into 3 mm disks in diameter. The disks were finally thinned using twin jet electropolishing with an electrolyte of 30% nitric acid and 70% methanol at 15 V at 25 C. Microstructure characterization was carried out by Tecnai F30 G2
Fig. 2 presents an EBSD band contrast map and a bright-field (BF) TEM image showing the typical microstructure of hot-rolled 7050 Al alloy which was solution treated at 475 C for 1 h and then water quenched to RT. One can see from the EBSD band contrast map in Fig. 2(a) that, although there exist elongated grains of 2e8 mm in width and 10e24 mm in length due to hot rolling, a lot of equiaxial grains of about 2e5 mm have been formed after the solution treatment. Fig. 2(b) shows grain size distribution histogram from Fig. 2(a). The grain equivalent diameter ranges from 2 mm to 17 mm. Fig. 2(c) shows some equiaxial grains of about 4 mm in diameter observed by TEM, and the grains have low dislocation density. 3.2. Microstructures after ECAP Fig. 3 presents the TEM micrographs with low and high magnification, revealing the microstructure after ECAP at RT for one pass. The subgrains were apparently elongated along the shear direction. The size of the elongated grains ranges from 100 to 500 nm in the longitudinal direction, and is about 100 nm in width, as shown in Fig. 3(a). High density of dislocations was homogeneously generated in the ECAPed alloy, as shown in Fig. 3(b). Fig. 3(c) is a typical HRTEM image along <011>Al direction showing the morphological and crystallographic features of the GPII zones [13]. Some planar precipitates are fully coherent with the Al matrix, parallel to {111}Al planes, with the thickness of about 1e3 times of {111}Al atomic plane spacing and the length of about 4e7 nm. Additionally, weak diffraction streaks are observed along {111}Al in the Fast Fourier transformation (FFT) spectrum of Fig. 3(c), indicating the GPII zones formed during ECAP processing. 3.3. Microstructures of the samples after aging treatment Fig. 4 shows the characterization of the precipitate microstructure of the unECAPed and the ECAPed samples aged at 160 C for 6 h Fig. 4(a) and (b) present BF TEM images of the two samples. In the unECAPed sample, fine dispersed disk-shaped precipitates with
Fig. 2. EBSD band contrast map (a), grain size distribution histogram (b) and BF TEM micrograph (c), showing the grains of the 7050 Al alloy after solid solution treatment.
M.H. Li et al. / Intermetallics 55 (2014) 49e55
51
Fig. 3. TEM micrographs of 7050 Al alloy after ECAP at RT for one pass showing the elongated grains (a), high density of dislocations (b), and some GPII zones in Al matrix from HRTEM micrograph of [011]Al projection (c). The inset is the corresponding FFT spectrum.
the size of about 2e6 nm in diameter are homogeneously distributed (Fig. 4(a)). To observe clearly the precipitates in the ECAPed sample, the contrast of dislocations are avoided by tilting the sample holder in Fig. 4(b). In the ECAPed sample, the large lathshaped precipitates have length of about 15e33 nm and thickness of about 7e16 nm, and the small disk-shaped precipitates have the size of about 3e10 nm in diameter. Fig. 4(c) and (d) shows typical <112>Al HRTEM images taken from the two samples. Fig. 4(c) shows some fine h0 precipitates (marked by a white arrow) in the unECAPed sample. The existence of the h0 phase is confirmed by the weak diffraction streaks at 1/3 and 2/3 {220}Al position in the FFT spectrum. These precipitates are about 5e8 nm long and about 2e4 nm thick. However, in the ECAPed sample, HRTEM image indicates that two types of precipitates coexist, namely h0 (marked by a white arrow) and h (marked by a black arrow) (Fig. 4(d)). Separate diffuse streaks and diffraction spots at 1/3 and 2/3 {220}Al position can be observed in the FFT spectrum, indicating the present of h0 and h phases. In this situation, the h0 precipitates have a length of about 4e7 nm and a thickness of about 2e3 nm, and h precipitates are about 15e20 nm long and about 4e7 nm thick. Fig. 4(e) and (f) shows the precipitate size histograms from Fig. 4(a) and (b) of TEM images, respectively. Two important phenomena can be observed from Fig. 4(e) and (f). First, the average precipitate equivalent diameter in the ECAPed sample is larger than that in the unECAPed sample after the same aging treatment (i.e., for 7.1 nm vs 4.0 nm). Second, the precipitate size distribution
widens in the ECAPed sample. In the unECAPed sample, the precipitate equivalent diameter ranges from 1 nm to 7 nm. However, in the ECAPed sample, the precipitate equivalent diameter ranges from 3 nm to 23 nm. Fig. 5(a) and (b) shows <011>Al selected area electron diffractions (SAEDs) patterns taken from the unECAPed and the ECAPed samples aged at 160 C for 6 h, respectively. Weak diffraction spot at the 1/3 and 2/3 {220}Al position (marked by a white arrow) indicates that h0 precipitates have formed in the unECAPed sample (Fig. 5(a)). However, two separate diffraction spots appears respectively at 1/3 and 2/3 {022}Al position (marked by a white arrow and a black arrow) in Fig. 5(b), indicating that h0 and h precipitates coexist in the aged ECAPed sample. 4. Discussion 4.1. Influence of temperature rise on precipitation According to Refs. [14e17], plastic deformation can lead to the temperature rise of the workpiece because of the heat generated by mechanical work and the friction between the ECAP die and the workpiece. Kim [16] proposed an equation for the temperature rise DT of the workpiece during ECAP processing:
DT ¼
.pffiffiffi 0:9s3 þ 0:5m s 3 uðA=VÞDt rC þ ðA=VÞhDt
(1)
52
M.H. Li et al. / Intermetallics 55 (2014) 49e55
Fig. 4. Characterization of the precipitate microstructure of the unECAPed sample (a, c, e) and the ECAPed sample (b, d, f) aged at 160 C for 6 h, showing BF TEM micrographs (a, b), HRTEM micrographs (c, d) and precipitate size histograms (e, f). The insets are the corresponding FFT spectrums.
where DT is the temperature increment of the workpiece in the main deforming zone, s is the stress, 3 is the strain imposed after pressing, m is the friction factor, u is the relative velocity between the die and the workpiece, h is the heat transfer coefficient between the workpiece and the die, A is the outer surface area of the domain contacting the die, V is the volume of the main deforming zone, Dt is the dwell time of the domain within the deforming zone, r is the density of the workpiece, and C is its heat capacity, respectively. When a sample is pressed through the die during ECAP processing, a strain, 3 , is introduced [18]:
3
¼N
2 cotðf=2 þ j=2Þ þ jcosecðf=2 þ j=2Þ pffiffiffi 3
(2)
where 3 is the strain imposed after multiple pressing, N is the number of pressing passes. For the present experiment, the sample was pressed using a die of F ¼ 90 , J ¼ 20 for one pass to give a strain 3 ¼ 1.0546. The average dwell time (Dt) of the domain within the deforming zone is:
M.H. Li et al. / Intermetallics 55 (2014) 49e55
53
Fig. 5. SAED patterns for samples aged at 160 C for 6 h: (a) the unECAPed sample and (b) the ECAPed sample. The inset is magnification of the rectangular region in (b).
Dt ¼
d j pffiffiffi u 2p
(3)
where d is the diameter of the cylindrical workpiece. For the present experiment, the cuboid workpiece with the size of 20 mm 20 mm 60 mm can be approximated as the cylindrical workpiece with a diameter 22.57 mm and 60 mm in length. u ¼ 6 mm/s. So Dt is evaluated to be 0.3 s. The parameter values for 7050 Al alloy, s ¼ 400 MPa, 3 ¼ 1.0546, r ¼ 2830 kg/m3, C ¼ 860 J/Kg K, m ¼ 0.2, and h ¼ 2000 N/m s K are used for calculation. By substituting these values into equation (1), DT ¼ 151 C can be finally obtained. Based on Refs. [14e17], there are a good agreement between the calculated, the measured and the simulated temperature rise of Al and Al alloys during ECAP processing. So, in the study, it is credible that the temperature is increased abruptly to 151 C in the 7050 Al alloy immediately upon passing through the shearing plane. Moreover, as Al alloy is a good thermal conductor, the temperature of the workpiece increases rapidly from the initial 25 C to a peak value of about 151 C, then decreases to 25 C in a few minutes during ECAP processing. Since the temperature rise may reach 151 C in 7050 Al alloy when the sample passed through the main deforming zone during ECAP processing, some GPII zones form due to the process of dynamic aging, as shown in Fig. 3(c). During post-ECAP artificial aging, the existing GPII zones can accelerate formation of metastable phase h0 . So, these GPII zones contribute to larger average precipitates size and the variation in precipitate size distribution in the post-ECAP aged sample. 4.2. Influence of vacancies on precipitation Solute segregation occurs along dislocations due to straininduced excess vacancies through the mechanism of nonequilibrium segregation [19,20]. The three necessary conditions for nonequilibrium segregation are 1) the existence of excess vacancies, 2) the presence of vacancy sinks, and 3) a positive vacancyimpurity binding energy. These three conditions are fully satisfied in the 7050 Al alloy subjected to ECAP, as discussed subsequently. 1) The existence of excess vacancies. SPD can create excess vacancies [21e23]. Deformation dependence of the vacancy concentration c of point defects may be written as [24]:
dc s ¼ d3 3m
(4)
where 3 is the strains, m is the shear modulus, s is the stress value and c is the vacancy concentration. This law is obeyed fairly well in face centered cubic metals [24]. Ref. [25] had supposed that the vacancy concentration is about 104 for strains 3 z 1, which is comparable to the equilibrium value at the melting point. 2) The presence of vacancy sinks. It is well known that dislocations are perfect vacancy sinks. ECAP can introduce high density of dislocations, as shown in Fig. 3(b). 3) A positive vacancy-impurity binding energy, Eb. The values of Eb can be calculated by using elasticity theory. According to the study of Cottrell [26], the following formula approximately describes the binding energy Eb of a vacancy with a foreign atom:
Eb ¼ 8pmr03 l2 ;
l ¼ ±ðr1 r0 Þ=r0
(5)
where l is the misfit of the foreign atom with matrix lattice, m is the shear modulus of the matrix, r0 is the matrix atom radius, and r1 is the foreign atom radius. Therefore, the binding energy of a vacancy with solute atom in 7050 Al alloy can be calculated as Mg Eb ¼ 0.28 eV, EbCu ¼ 0.22 eV and EbZn ¼ 0.06 eV. Based on the theory of nonequilibrium segregation, solute atoms gradually segregate to dislocations. Since dislocations serve as an ideal sink of vacancy, the level of excess vacancies at dislocation is zero. So, soluteevacancy complex gradients could develop near dislocations. Subsequently, soluteevacancy complexes diffuse towards dislocations due to the concentration gradients, and then the soluteevacancy complexes are decomposed into vacancies and solute atoms (Mg, Cu and Zn). Meanwhile, the vacancies will annihilate at dislocations leaving behind the solute atoms. As the increase of solute concentration at dislocations, a subsequent desegregation occurs. However, desegregation of solute is negligible because the diffusivity of single solute is much slower than that of soluteevacancy complex [19,27,28]. As a result, solute atoms gradually segregate to dislocations. 4.3. Influence of dislocations on precipitation ECAP introduces high density of dislocations both at the grain/ subgrain boundaries, as seen in Fig. 3(b). For Al alloy materials subjected to ECAP processing, dislocations density r is up to
54
M.H. Li et al. / Intermetallics 55 (2014) 49e55
1013~1014/m2 after one pass ECAP [5]. The increased dislocation density has two aspects to influence the precipitation reaction of 7050 A1 alloy. The presence of dislocations induces not only faster and coarser precipitation on dislocations but also a large precipitate size gradient. In the ECAPed sample, precipitation was strongly accelerated in aging process, as shown in Fig. 4(b). The average size of the precipitates in the ECAPed sample (Fig. 4(f)) is much larger than that in the unECAPed sample (Fig. 4(e)). According to the HRTEM images and the diffraction observed near 1/3 and 2/3 {220} in the SAEDs shown in Fig. 4(c, d) and 5, the metastable phases h0 and the equilibrium phases h coexist in the ECAPed sample whereas only the metastable phases h0 are present in the unECAPed sample aged at 160 C for 6 h. Moreover, the precipitate size distribution becomes wide in the ECAPed sample (see Fig. 4e, f). In the unECAPed sample, the precipitate equivalent diameter ranges from 1 nm to 7 nm. However, in the ECAPed sample, the precipitate equivalent diameter ranges from 3 to 23 nm. The two phenomena indicate that dislocations not only lead to nucleation and growth of larger precipitates on dislocations, but also modify homogeneous precipitation in the matrix surrounding dislocations. This can be explained from the following two aspects.
treatment is a complex precipitation process, since ECAP generates deformation heat, large quantities of vacancies and high density of dislocations. Some conclusions are summarized as follows:
1) Dislocations can reduce the activation energy for precipitation by reducing the strain energy or interface energy. Thus, dislocations act readily as heterogeneous nucleation sites [8]. Moreover, dislocations increase diffusivities of precipitateforming elements due to short circuit diffusion along dislocations [29], resulting in the rapid growth and coarsening of precipitates on dislocations.
Acknowledgments
At the early stages of artificial aging, the vacancyesolute complexes gradually migrated to dislocations due to nonequilibrium segregation. In the vicinity of dislocations, frequent collisions between these soluteevacancy complexes released vacancies which finally annihilated at dislocations leaving behind mobile complexes of solute atoms which may or may not also include vacancies. As a result, fine soluteerich clusters are formed at the beginning of artificial aging which act as GPII zones nuclei on subsequent aging. Subsequent evolution of the microstructure involves the replacement of the GPII zones with more stable phases. 2) The influence of dislocation on precipitation can't just be considered from the heterogeneous precipitation on dislocations, as described in Section 4.2. On the one hand, large quantities of solutes flow to dislocations by the mechanism of nonequilibrium segregation [19,20]. The solute concentration gradient forms around the dislocations. On the other hand, dislocations are also ideal vacancy sinks [19,20,30]. Therefore, there exists the vacancy concentration gradient around the dislocations. Since the concentration and distribution of vacancy have the strong influence on the homogeneous nucleation of precipitates in AleZneMgeCu alloys [31,32], the presence of high density of dislocations will influence homogeneous precipitation around them. As a result, the flow of solutes and vacancies to dislocations leads to competition between homogeneous precipitation in matrix and heterogeneous precipitation on dislocations, and subsequently a large precipitate size gradient has been formed. 5. Conclusions The influence of ECAP at RT on aging precipitation of 7050 Al alloy was investigated by using TEM and HRTEM. Post-ECAP aging
1) Due to deformation heat induced by SPD, the temperature rise may reach 151 C when the sample passes through the main deforming zone during ECAP processing. Some GPII zones can be observed after ECAP, which will accelerate precipitation during subsequent artificial aging, and induce the variation in precipitate size distribution. 2) High density of dislocations can accelerate nucleation as well as growth of precipitate phases. This leads to coarser precipitation on dislocations in the process of post-ECAP aging in 7050 Al alloy. 3) Although the effect of ECAP on aging precipitation is mainly heterogeneous precipitation on dislocations, due to the flow of solutes and vacancies to dislocations by the mechanism of nonequilibrium segregation, there exists competition between homogeneous precipitation in matrix and heterogeneous precipitation on dislocations. As a result, the precipitate size distribution widens.
The authors would like to acknowledge the financial support of the 111 Project (B08040) of China and the Natural Science Foundation of China (51071125, 51271147 and 51201135).
References [1] Valiev RZ, Langdon TG. Principles of equal-channel angular pressing as a processing tool for grain refinement. Prog Mater Sci 2006;51:881e981. [2] Valiev RZ, Islamgaliev RK, Alexandrov IV. Bulk nanostructured materials from severe plastic deformation. Prog Mater Sci 2000;45:103e89. [3] Kim WJ, Kim JK, Kim HK, Park JW, Jeong YH. Effect of post equal-channelangular-pressing aging on the modified 7075 Al alloy containing Sc. J Alloys Compd 2008;450:222e8. [4] Kim WJ, Chung CS, Ma DS, Hong SI, Kim HK. Optimization of strength and ductility of 2024 Al by equal channel angular pressing (ECAP) and post-ECAP aging. Scr Mater 2003;49:333e8. [5] Zhao YH, Liao XZ, Jin Z, Valiev RZ, Zhu YT. Microstructures and mechanical properties of ultrafine grained 7075 Al alloy processed by ECAP and their evolutions during annealing. Acta Mater 2004;52: 4589e99. [6] Zhao YH, Liao XZ, Cheng S, Ma E, Zhu YT. Simultaneously increasing the ductility and strength of nanostructured alloys. Adv Mater 2006;18: 2280e3. [7] Horita Z, Ohashi K, Fujita T, Kaneko K, Langdon TG. Achieving high strength and high ductility in precipitation-hardened alloys. Adv Mater 2005;17: 1599e602. [8] Gubicza J, Schiller I, Chinh NQ, Illy J, Horita Z, Langdon TG. The effect of severe plastic deformation on precipitation in supersaturated AleZneMg alloys. Mater Sci Eng A 2007;460e461:77e85. chet Y, Chemin JL, Hutchinson CR. In situ eval[9] Deschamps A, Fribourg G, Bre uation of dynamic precipitation during plastic straining of an AleZneMgeCu alloy. Acta Mater 2012;60:1905e16. [10] Deschamps A, De Geuser F, Horita Z, Lee S, Renou G. Precipitation kinetics in a severely plastically deformed 7075 aluminium alloy. Acta Mater 2014;66: 105e17. [11] Han WZ, Chen Y, Vinogradov A, Hutchinson CR. Dynamic precipitation during cyclic deformation of an underaged AleCu alloy. Mater Sci Eng A 2011;528: 7410e6. [12] Wolverton C. Soluteevacancy binding in aluminum. Acta Mater 2007;55: 5867e7872. [13] Berg LK, Gjønnes J, Hansen V, Li XZ, Knutson-Wedel M, Waterloo G, et al. GPzones in AleZneMg alloys and their role in artificial aging. Acta Mater 2001;49:3443e51. [14] Yamaguchi D, Horita Z, Nemoto M, Langdon TG. Significance of adiabatic heating in equal-channel angular pressing. Scr Mater 1999;41: 791e6. [15] Pei QX, Hu BH, Lu C, Wang YY. A finite element study of the temperature rise during equal channel angular pressing. Scr Mater 2003;49:303e8. [16] Kim HS. Prediction of temperature rise in equal channel angular pressing. Mater Trans 2001;42:536e8.
M.H. Li et al. / Intermetallics 55 (2014) 49e55 [17] Hosford WF, Caddell RM. Metal forming: mechanics and metallurgy. 3rd ed. New York: Cambridge University Press; 2007. [18] Iwahashi Y, Wang J, Horita Z, Nemoto M, Langdon TG. Principle of equalchannel angular pressing for the processing of ultra-fine grained materials. Scr Mater 1996;35:143e6. [19] Liu WJ. A new theory and kinetic modeling of precipitation of Nb(CN) in microalloyed strain-induced austenite. Metall Mater Trans A 1995;26: 1641e57. [20] Liu WC, Chen ZL, Yao M. Effect of cold rolling on the precipitation behavior of d phase in inconel 718. Metall Mater Trans A 1999;30:31e40. [21] Straumal BB, Baretzky B, Mazilkin AA, Phillipp F, Kogtenkova OA, Volkovb MN, et al. Formation of nanograined structure and decomposition of supersaturated solid solution during high pressure torsion of AleZn and AleMg alloys. Acta Mater 2004;52:4469e78. [22] Sauvage X, Wetscher F, Pareige P. Mechanical alloying of Cu and Fe induced by severe plastic deformation of a CueFe composite. Acta Mater 2005;53: 2127e35. [23] Su LH, Lu C, He LZ, Zhang LC, Guagliardo P, Tieu AK, et al. Study of vacancytype defects by positron annihilation in ultrafine-grained aluminum severely deformed at room and cryogenic temperatures. Acta Mater 2012;60: 4218e28.
55
[24] Saada G. Interaction of dislocations and hardening production of point defects in fcc metals. Acta Metall 1961;9:166e8. [25] Friedel J. Dislocations. Oxford: Pergamon Press; 1964. [26] Cottrell AH. An introduction to metallurgy. London: Edward Arnold; 1967. [27] Xu TD, Song SH, Shi HZ, Gust W, Yuan ZX. A method of determining the diffusion-coefficient of vacancy solute atom complexes during the segregation to grain-boundaries. Acta Metall Mater 1991;39:3119e24. [28] Williams TM, Stoneham AM, Harries DR. The segregation of boron to grain boundaries in solution-treated Type 316 austenitic stainless steel. Met Sci 1976;10:14e9. [29] Hoyt JJ. On the coarsening of precipitates located on grain boundaries and dislocations. Acta Metall 1991;39:2091e8. chet Y. Influence of predeformation on ageing in an [30] Deschamps A, Livet F, Bre AleZneMg alloydI. Microstructure evolution and mechanical properties. Acta Mater 1999;47:281e92. [31] Girifalco LA, Herman H. A model for the growth of GuinierePreston zonesdthe vacancy pump. Acta Metal 1965;13:583e90. [32] Embury JD, Nicholson RB. The nucleation of precipitates: the system AleZneMg. Acta Metall 1965;13:403e17.