- Email: [email protected]
Effects of deep cryogenic treatment on the microstructure and mechanical properties of commercial pure zirconium
Journal of Alloys and Compounds 619 (2015) 513–519
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Effects of deep cryogenic treatment on the microstructure and mechanical properties of commercial pure zirconium Chao Yuan, Yunpeng Wang, Deli Sang, Yijun Li, Lei Jing, Ruidong Fu ⇑, Xiangyi Zhang State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, PR China College of Materials Science and Engineering, Yanshan University, Qinhuangdao, Hebei 066004, PR China
a r t i c l e
i n f o
Article history: Received 25 April 2014 Received in revised form 3 August 2014 Accepted 26 August 2014 Available online 16 September 2014 Keywords: Metals and alloys Deep cryogenic treatment Microstructure Mechanical properties Zirconium
a b s t r a c t The effects of deep cryogenic treatment (DCT) on the microstructure and mechanical properties of commercial pure zirconium were investigated. Experimental results indicated that DCT induced a change in grain orientation and improved internal stress, which in turn increased dislocation density that led to improved hardness. Hardness in basal planes was found to be significantly larger than that in prism planes. Moreover, strength was enhanced in DCT-treated zirconium and the ductility was comparable to that of as-annealed zirconium. This phenomenon was due to the increase in dislocation density and the good ductility resulting from the motion of pre-existing dislocations and specific dislocation configurations. DCT led to the transformation of tensile fracture mode from mixed-rupture characteristics of quasi-cleavage and dimples to quasi-cleavage, thereby increasing compatible deformation capabilities. The possible mechanisms underlying microstructural modification, tensile strength, and hardness improvement were discussed. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Cryogenic treatment is a very old process that is widely used for high-precision components. Shallow cryogenic treatment is set at low temperatures (about 80 °C), whereas deep cryogenic treatment (DCT) is set at near-liquid-nitrogen temperatures (about 196 °C). DCT improves certain properties beyond the enhancement obtained by normal cold treatment [1–3]. DCT is different from those methods of severe plastic deformation (SPD) processing at cryogenic conditions, such as cryogenic rolling, pinning and milling. Instead of nanostructured and ultrafine-grains in SPD-processed metals, the variation of crystal defect and phase transformation may play more important role in increasing of the properties of DCT-treated metals. Over the past few decades, the effects of various cryogenic treatments on the performance of steel have received considerable interest. DCT reportedly increases the normal temperature strength and hardness of steels [4,5], provides dimensional stability or microstructural stability [6], and improves wear [7–9] and fatigue resistance [10,11]. The improvement in mechanical properties can be ascribed to the complete transformation of retained austenite into martensite, precipitation of fine dis⇑ Corresponding author at: State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, PR China. Tel.: +86 335 858 7046; fax: +86 335 807 4545. E-mail address: [email protected] (R. Fu). http://dx.doi.org/10.1016/j.jallcom.2014.08.201 0925-8388/Ó 2014 Elsevier B.V. All rights reserved.
persed carbides, and removal of residual stresses [6,7,12]. Previous studies have shown that the hardness and abrasion resistance of DCT-treated samples evidently improve because of transformation of abundant retained austenite to martensite, secondary carbide precipitation [13], and the precipitation of nanosized g-carbides in primary martensite [14]. In addition, DCT combined with tempering treatment enables the enhancement of fatigue properties for the precipitation of fine carbides and the reduction in compressive residual stress [15]. Compared with studies on ferrous metals, research on the effect of DCT on nonferrous metals such as Mg, Al, and Ti alloys are limited. Asl et al. [2] found that after DCT, tiny laminar b phase particles almost dissolve, and the coarse divorced eutectic b phase extends into the neighboring matrix. As a result, the mechanical properties of AZ91 Mg alloy significantly improve. Jiang et al. [16] found that DCT induces the refinement of grains to approximately 0.1–3.0 lm, thereby improving the strength and elongation of 3102 Al alloy foil. As aforementioned, evident changes in microstructure such as phase transformation or grain refinement have been verified for DCT-treated alloys. However, changes in microstructure and mechanical properties that can be introduced by DCT if no phase transformation occurs are worth investigating. In this study, coarse-grained zirconium (Zr) with a hexagonal close packed (hcp) structure and higher microstructure stability at cryogenic
514
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
temperature was chosen as a model metal because of the limited deformation capability that results in easily saturated dislocation density [17]. Some unusual changes in mechanical properties were found in DCT-treated Zr, and the possible mechanisms in terms of microstructural modification were discussed. 2. Experimental The material used in this study was commercial pure Zr plate (chemical composition summarized in Table 1). The plate was annealed in a vacuum at 1123 K for 4 h to obtain a homogeneous polycrystalline structure. The average grain size of the as-annealed sample was approximately 30–50 lm. The samples cut from the as-annealed plate were placed in a liquid nitrogen environment at 77 K for 24 h. Afterwards, the samples were taken out and cooled to room temperature. X-ray diffraction (XRD) patterns of the as-annealed and DCT-treated samples were measured using a Rigaku D/max 2500 X-ray diffractometer (18 kW) with Cu Ka radiation (wavelength k = 1.54 Å) in continuous-scanning mode over 2h = 30– 80° at a step of 0.02°. Differential scanning calorimetry (DSC) was conducted on a Diamond DSC (PerkinElmer Inc., UK). For dynamic scans, samples were subjected to temperatures ranging from 20 °C to 160 °C at a cooling rate of 20 °C/min. Microstructure features before and after DCT were observed by optical microscopy (OM), electron backscatter diffraction (EBSD), and transmission electron microscopy (TEM). EBSD analysis was performed using a HITACHI S-4800 scanning electron microscopy (SEM) system, whereas TEM observations were carried out with a JEOL-2010 TEM system at a voltage of 200 kV. A FM-ARS9000 Vickers microhardness tester was used to measure the hardness of samples. Indentations were carried out with a given load of 200 g and a dwell time of 10 s. Hardness distributions over a square area of 4.5 4.5 mm2, with a space between adjacent indentations of 0.5 mm were obtained. The tensile samples were machined with a gauge length of 30 mm and a cross-section of 10 1.5 mm2. Four repeated tensile tests were performed on a MTS test system at a strain rate of 1 104 s1 at room temperature under extensometer-measured strain control. Fracture surfaces after the tensile test were observed by SEM, and microstructures near the fracture surfaces were observed by TEM.
3. Results and discussion 3.1. XRD and DSC analyses Fig. 1(a) shows the XRD patterns of as-annealed and DCTtreated samples. No phase transformation was observed in DCTtreated Zr compared with as-annealed Zr. However, the intensities of the diffraction peaks of DCT-treated Zr were slightly enhanced, which can be attributed to the variation in lattice constants caused by DCT. Changes in lattice constants were also observed in DCTtreated Mg [18,19]. To further confirm the microstructure stability of Zr at cryogenic temperature, a cryogenic DSC curve was obtained and is shown in Fig. 1(b). The smooth feature of the DSC curve indicated that no transformation occurred during DCT. 3.2. EBSD analysis To gain deep insight into microstructural changes, EBSD analysis was performed. In this analysis, orientation imaging microscopy (OIM) maps, inverse pole figure (IPF) maps, and boundary misorientation angle distributions were obtained before and after DCT, as shown in Fig. 2. Fig. 2(a) and (b) shows the OIM maps of Zr before and after DCT, with high-angle (HAGBs; grain boundary misorientations P 15°) and low-angle (LAGBs; grain boundary misorientations < 15°) grain boundaries depicted by black and white lines, respectively. The crystallographic directions corresponding to various colors can be inferred from the IPF triangle shown at the bottom right corner of Fig. 2(a) and (b). The microstructure was characterized by equiaxed grains and a higher
Fig. 1. XRD patterns of as-annealed and DCT-treated samples (a), and the cryogenic DSC curve of Zr (b).
fraction of HAGBs. No phase transformation and precipitation occurred during DCT, consistent with the XRD and DSC results. Fig. 2(c) and (d) shows the IPF maps of Zr before and after DCT, respectively. The maximum intensity indicated the number of random orientations. A strong preference was observed toward the (0 0 0 1) orientation, which was corroborated by the OIM maps in Fig. 2(a) and (b). Furthermore, the intensity of prism planes was lower and the grain orientations were much closer to the (0 0 0 1) basal plane orientation after DCT. This finding can be attributed to the ordered arrangement of atoms resulting from the small changes in lattice constants [18,19]. The misorientation angle distributions in Fig. 2(e) and (f) shows that the crystallite boundaries were mainly high angle in nature. The HAGB fractions of Zr before and after DCT were about 81.2% and 85.3% of the total grain boundary length, respectively. Thus, DCT improved the formation of HAGBs. The increase in HAGB fractions may be related to the extrinsic dislocations formed during DCT (Fig. 3). The extrinsic dislocations were probably easier to react with the dislocations in LAGBs [20,21]. Therefore, the accumulation and rearrangement of dislocations resulted in the transformation of LAGBs to HAGBs, leading to increased HAGB and decreased LAGB fractions. 3.3. TEM observations
Table 1 Chemical composition (in wt%) of the investigated pure Zr. Fe + Cr
C
N
H
O
Hf
Zr
0.2
0.05
0.01
0.005
0.16
4.5
Balance
The TEM images of the samples before and after DCT are presented in Fig. 3. The dislocation density of the DCT-treated sample was apparently higher than that of the as-annealed sample. When the temperature dropped from room temperature to cryogenic
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
515
Fig. 2. Representative OIM maps of (a) as-annealed and (b) DCT-treated samples. IPF maps of (c) as-annealed and (d) DCT-treated samples. Distribution of boundary misorientation angle of (e) as-annealed and (f) DCT-treated samples.
temperature, the volume of the material contracted. This volume contraction released great compression deformation energy that served as the driving force for the formation and movement of dislocations [18]. The improvement in dislocation density played a significant role in enhancing mechanical properties and deformation behavior, as discussed in the following sections. 3.4. Hardness measurements To better understand the variation in hardness of Zr, contour maps of the spatial distribution of the Vickers hardness (Hv) over a 4.5 4.5 mm2 square area before and after DCT are shown in Fig. 4(a) and (b), respectively. The hardness of as-annealed Zr
substantially varied from approximately 140–190 Hv, with an average value of 159.5 Hv. By contrast, the hardness of DCT-treated Zr generally changed from approximately 160–220 Hv, with an average value of 197.5 Hv. Evidently, the average hardness significantly increased after DCT. The difference between the two average hardness values reached 23.8%. Although the spatial distribution of hardness was not uniform from one area to another, the increase in hardness of the DCT-treated sample was a holistic rather than localized phenomenon, as can also be confirmed by the hardness evolution characterized by the color maps in Fig. 4. The inhomogeneity of hardness distribution may be closely related to the grain orientation. An in situ comparison between the OIM map and OM image of hardness indentation of DCT-treated Zr is presented
516
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
Fig. 3. TEM images of (a) as-annealed and (b) DCT-treated samples.
among different grain orientations reached up to 30.2%. The dependency of hardness values on orientation has also been observed in previous reports [22–24]. For hcp Zr, the axial ratio c/a was 1.593, which was less than that of the ideal value (i.e., <1.633). Consequently, the lattice resistance for prism planes was lower than that for basal or pyramidal planes; thus, slip preferred to occur on prism planes [25]. In other words, plastic deformation more easily occurred on prism planes than on basal planes. Consequently, the hardness in basal planes was much larger than that in prism planes. Therefore, the overall improvement of hardness may be closely related to both changes in grain orientation and dislocation density. After DCT, the grain orientation tended to be more consistent, resulting in ordered strengthening [18]. Meanwhile, the increase in dislocation density increased the force to overcome the resistance of dislocation motion. Thus, the improvement in plastic deformation resistance resulted in increased hardness. 3.5. Tensile properties The tensile engineering and true stress–strain curves of as-annealed and DCT-treated Zr are presented in Fig. 6. For as-annealed Zr, the yield strength (0.2% offset, r0.2) and ultimate tensile strength (rUTS) were 288 and 393 MPa, respectively. The elongation to fracture reached 28.7%. Correspondingly, r0.2 and rUTS of DCT-treated Zr increased by about 60% and 40%, respectively, compared with as-annealed Zr. Meanwhile, the elongation to fracture reached 26.5%, which was similar to that of as-annealed Zr. This finding indicated that DCT was a better route to improving the strength and keeping a good ductility of Zr. As depicted in Fig. 6(b), DCT-treated Zr also exhibited high strain hardening capabilities relative to as-annealed Zr. The strain hardening exponent (n) is defined by [26]
r ¼ K en
Fig. 4. Spatial distributions of the hardness values of Zr: (a) before and (b) after DCT.
in Fig. 5. The letters in Fig. 5(a) correspond to those in Fig. 5(b), and grain M indicates the reference grain. The hardness values of grains A–D in Fig. 5(b) were 211.65, 259.54, 199.35, and 233.57 Hv, respectively. The hardness values varied with the orientation between basal and prism planes. Planes close to the (0 0 0 1) basal plane orientation were significantly harder than planes close to 0Þ and ð2 1 1 0Þ planes. The hardness difference the prism ð1 0 1
ð1Þ
where r is the true stress, e is the true strain, and K is a constant. The n values for as-annealed and DCT-treated Zr were 0.09 and 0.11, respectively. The enhancement in strain hardening capabilities of DCT-treated Zr was further confirmed by an improved strain hardening rate (H) defined by [26]
H¼
1 @r r @e
ð2Þ
Variations in H with e before and after DCT are displayed in Fig. 7. The variation trends were consistent with continuous strain hardening to significant strains, but the strain hardening rate of the DCT-treated sample was larger than that of the as-annealed sample.
517
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
(a)
M
A
(b)
D B
C
B
C D A
M
Fig. 5. (a) OIM map and (b) corresponding OM image of hardness indentation of the DCT-treated Zr.
Fig. 7. Normalized strain hardening rate (H) against the true strain of as-annealed and DCT-treated samples.
Fig. 6. Tensile (a) engineering and (b) true stress–strain curves of as-annealed and DCT-treated samples.
DCT-treated Zr exhibited enhancement of strength and maintained the good ductility compared with as-annealed Zr. The variation in mechanical properties can be attributed to the difference in dislocation density and dislocation configurations produced during deformation [17,27,28]. To reveal the mechanism underlying the increase in strength and maintain the good ductility in DCTtreated Zr, TEM observations of the fractured tensile samples were performed (Fig. 8). As shown in Fig. 8(a), for as-annealed Zr after tensile deformation, the grains were apparently distorted and elongated, and masses of dislocations were found in the grains. However, the deformation of DCT-treated Zr was more uniform and
compatible than that of as-annealed Zr. Many dislocations were also observed within the grains, which were divided into smaller blocks (Fig. 8(b)). During DCT, a large number of pre-existing dislocations were introduced into the grains (Fig. 3(b)), and these grains were movable during tensile deformation [27–32]. At the initial stage of deformation, more dislocations and dislocation tangles formed. The increase in dislocation density and resistance of dislocation motion resulted in improved strength. With increased strain, many irregular dislocation tangles changed into parallel dislocation lamellas along with the tensile direction. With further increased strain, dislocation accumulation and rearrangement occurred to form dislocation cells, dense dislocation walls, and nanoscale subgrains and grains. As a result, the coarse grains were divided into smaller blocks. These blocks made the plastic deformation of DCT-treated Zr more uniform and compatible, thereby resulting in the ductility being comparable to that of as-annealed Zr. 3.6. Fracture behavior The fracture surfaces of the as-annealed and DCT-treated Zr sample are shown in Fig. 9. As shown in Fig. 9(a), the fracture surface of edge regions of the as-annealed sample exhibited mixed characteristics with dimples and a small amount of quasi-cleavage facets, whereas the fracture surface in the middle region as shown in Fig. 9(b) mainly contained many quasi-cleavage facets. The dimples in the edge of fracture surface of as-annealed sample were related to the noticeable necking caused by severe strain
518
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
Fig. 8. TEM images of (a) as-annealed and (b) DCT-treated samples after tensile tests.
(a)
(b)
500 μm
500 μm
(c)
(d)
500 μm
500 μm
Fig. 9. SEM images showing the fracture surfaces of as-annealed and DCT-treated Zr tensile samples. (a) and (b) are the edge and middle region of as-annealed Zr sample; (c) and (d) are the edge and middle region of DCT-treated Zr sample.
localization during tensile test. For comparison, the fracture surfaces of DCT-treated Zr in the same regions are presented in Fig. 9(c) and (d), respectively. The fracture characteristics of two regions were mainly consisted of quasi-cleavage facets. No obvious necking was observed in the edge and middle regions of the fracture surfaces. The above findings indicated that DCT restricted strain localization occurring in the edge regions of tensile samples. For asannealed Zr, plastic instability (i.e., necking) occurred in the edge region of tensile samples before ultimate fracture. Moreover, Fig. 2(e) and (f) shows that the fraction of LAGBs before DCT was slightly larger than that after DCT. LAGBs are reportedly associated with the early onset of plastic instability during tensile testing [33–35]. The plastic instability accounted for the different fracture
characteristics in the edge and middle region exhibited by asannealed Zr. By contrast, the motion of pre-existing dislocations and specific dislocation configurations effectively enhanced the compatible deformation capability, resulting in the same fracture characteristics in both regions of the DCT-treated sample. 4. Conclusion Based on the above-described investigations, the following conclusions were drawn. (1) The grain orientations were much closer to the (0 0 0 1) basal plane orientation after DCT. DCT increased HAGB fractions and dislocation density.
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
(2) Hardness was closely related to grain orientations and dislocation density. The hardness in basal planes was much higher than that in prism planes. After DCT, dislocation density increased the resistance of dislocation motion, thereby improving hardness. (3) Strength levels were high and good ductility could still be achieved after DCT. The enhancement in strength was due to the increase in dislocation density, and the maintenance of good ductility resulted from the motion of pre-existing dislocations and specific dislocation configurations. (4) The fracture surfaces exhibited mixed features with quasicleavage facets and dimples for as-annealed Zr and quasicleavage facets for DCT-treated Zr. The grains were divided into smaller blocks because of the interaction of dislocations during the tensile process of DCT-treated Zr. Consequently, the compatible deformation capability of materials was enhanced and good ductility could still be achieved.
Acknowledgment Financial support from the National Basic Research Program of China (Grant No. 2010CB731606) is greatly acknowledged. References [1] [2] [3] [4]
R.F. Barron, Cryogenics 22 (1982) 409–414. K.M. Asl, A. Tari, F. Khomamizadeh, Mater. Sci. Eng., A 523 (2009) 27–31. M. Koneshlou, K.M. Asl, F. Khomamizadeh, Cryogenics 51 (2011) 55–61. A. Molinari, M. Pellizzari, S. Gialanella, G. Straffelini, K.H. Stiasny, J. Mater. Process. Technol. 118 (2001) 350–355. [5] P. Baldissera, Mater. Des. 31 (2010) 4725–4730. [6] P. Baldissera, C. Delprete, Open Mech. Eng. J. 2 (2008) 1–11. [7] B. Podgornik, F. Majdic, V. Leskovsek, J. Vizintin, Wear 288 (2012) 88–93.
519
[8] R. Thornton, T. Slatter, A.H. Jones, R. Lewis, Wear 271 (2011) 2386–2395. [9] T. Slatter, R. Lewis, A.H. Jones, Wear 271 (2011) 1481–1489. [10] P.J. Singh, S.L. Mannan, T. Jayakumar, D.R.G. Achar, Eng. Fail. Anal. 12 (2005) 263–271. [11] S. Zhirafar, A. Rezaeian, M. Pugha, J. Mater. Process. Technol. 186 (2007) 298– 303. [12] V. Leskovšek, B. Podgornik, Mater. Sci. Eng., A 531 (2012) 119–129. [13] H.H. Liu, J. Wang, B.L. Shen, H.S. Yang, S.J. Gao, S.J. Huang, Mater. Des. 28 (2007) 1059–1064. [14] P.F. Stratton, Mater. Sci. Eng., A 449 (2007) 809–812. [15] A. Bensely, S. Venkatesh, D. Mohan Lal, G. Nagarajan, A. Rajadurai, Krzysztof Junik, Mater. Sci. Eng., A 479 (2008) 229–235. [16] X.Q. Jiang, N. Li, H. He, X.J. Zhang, C.C. Li, H. Yang, Mater. Sci. Forum 546–549 (2007) 845–848. [17] D.F. Guo, M. Li, Y.D. Shi, T.Y. Ma, Z.B. Zhang, T.S. Wang, X.Y. Zhang, Mater. Sci. Eng., A 581 (2013) 133–139. [18] J.W. Liu, G.F. Li, D. Chen, Z.H. Chen, Chin. J. Aeronaut. 25 (2012) 931–936. [19] X.F. Zhang, G.H. Wu, W.C. Liu, W.J. Ding, Trans. Nonferrous Met. Soc. China 22 (2012) 2883–2890. [20] P. Xue, B.L. Xiao, Z.Y. Ma, Mater. Sci. Eng., A 532 (2012) 106–110. [21] Y.H. Zhao, J.F. Bingert, Y.T. Zhu, X.Z. Liao, R.Z. Valiev, Z. Horita, T.G. Langdon, Y.Z. Zhou, E.J. Lavernia, Appl. Phys. Lett. 92 (2008) 081903. [22] D.N. French, D.A. Thomas, Trans. AIME 233 (1965) 950–952. [23] T. Takahashi, E.J. Freise, Philos. Mag. 12 (1964) 1–8. [24] B. Roebuck, P. Klose, K.P. Mingard, Acta Mater. 60 (2012) 6131–6143. [25] L. Zhang, Y. Han, J. Lu, Nanotechnology 19 (2008) 165706. [26] G.E. Dieter, Mech. Metall., third ed., McGraw-Hill, London, 1988. [27] D.F. Guo, M. Li, Y.D. Shi, Z.B. Zhang, T.Y. Ma, H.T. Zhang, X.Y. Zhang, Mater. Sci. Eng., A 558 (2012) 611–615. [28] D.F. Guo, M. Li, Y.D. Shi, Z.B. Zhang, H.T. Zhang, X.M. Liu, X.Y. Zhang, Mater. Lett. 66 (2012) 305–307. [29] X.X. Huang, Scr. Mater. 60 (2009) 1078–1082. [30] L. Lu, X. Chen, X. Huang, K. Lu, Science 323 (2009) 607–610. [31] X.X. Huang, N. Hansen, N. Tsuji, Science 312 (2006) 249–251. [32] Z.W. Shan, Raja K. Mishra, S.A. Syed Asif, Oden L. Warren, Andrew M. Minor, Nat. Mater. 7 (2008) 115–119. [33] R. Kapoor, N. Kumar, R.S. Mishra, C.S. Huskamp, K.K. Sankaran, Mater. Sci. Eng., A 527 (2010) 5246–5254. [34] P.L. Sun, C.Y. Yu, P.W. Kao, C.P. Chang, Scr. Mater. 52 (2005) 265–269. [35] N. Kumar, R.S. Mishra, C.S. Huskamp, K.K. Sankaran, Mater. Sci. Eng., A 528 (2011) 5883–5887.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Effects of deep cryogenic treatment on the microstructure and mechanical properties of commercial pure zirconium Chao Yuan, Yunpeng Wang, Deli Sang, Yijun Li, Lei Jing, Ruidong Fu ⇑, Xiangyi Zhang State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, PR China College of Materials Science and Engineering, Yanshan University, Qinhuangdao, Hebei 066004, PR China
a r t i c l e
i n f o
Article history: Received 25 April 2014 Received in revised form 3 August 2014 Accepted 26 August 2014 Available online 16 September 2014 Keywords: Metals and alloys Deep cryogenic treatment Microstructure Mechanical properties Zirconium
a b s t r a c t The effects of deep cryogenic treatment (DCT) on the microstructure and mechanical properties of commercial pure zirconium were investigated. Experimental results indicated that DCT induced a change in grain orientation and improved internal stress, which in turn increased dislocation density that led to improved hardness. Hardness in basal planes was found to be significantly larger than that in prism planes. Moreover, strength was enhanced in DCT-treated zirconium and the ductility was comparable to that of as-annealed zirconium. This phenomenon was due to the increase in dislocation density and the good ductility resulting from the motion of pre-existing dislocations and specific dislocation configurations. DCT led to the transformation of tensile fracture mode from mixed-rupture characteristics of quasi-cleavage and dimples to quasi-cleavage, thereby increasing compatible deformation capabilities. The possible mechanisms underlying microstructural modification, tensile strength, and hardness improvement were discussed. Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction Cryogenic treatment is a very old process that is widely used for high-precision components. Shallow cryogenic treatment is set at low temperatures (about 80 °C), whereas deep cryogenic treatment (DCT) is set at near-liquid-nitrogen temperatures (about 196 °C). DCT improves certain properties beyond the enhancement obtained by normal cold treatment [1–3]. DCT is different from those methods of severe plastic deformation (SPD) processing at cryogenic conditions, such as cryogenic rolling, pinning and milling. Instead of nanostructured and ultrafine-grains in SPD-processed metals, the variation of crystal defect and phase transformation may play more important role in increasing of the properties of DCT-treated metals. Over the past few decades, the effects of various cryogenic treatments on the performance of steel have received considerable interest. DCT reportedly increases the normal temperature strength and hardness of steels [4,5], provides dimensional stability or microstructural stability [6], and improves wear [7–9] and fatigue resistance [10,11]. The improvement in mechanical properties can be ascribed to the complete transformation of retained austenite into martensite, precipitation of fine dis⇑ Corresponding author at: State Key Laboratory of Metastable Materials Science and Technology, Yanshan University, Qinhuangdao, Hebei 066004, PR China. Tel.: +86 335 858 7046; fax: +86 335 807 4545. E-mail address: [email protected] (R. Fu). http://dx.doi.org/10.1016/j.jallcom.2014.08.201 0925-8388/Ó 2014 Elsevier B.V. All rights reserved.
persed carbides, and removal of residual stresses [6,7,12]. Previous studies have shown that the hardness and abrasion resistance of DCT-treated samples evidently improve because of transformation of abundant retained austenite to martensite, secondary carbide precipitation [13], and the precipitation of nanosized g-carbides in primary martensite [14]. In addition, DCT combined with tempering treatment enables the enhancement of fatigue properties for the precipitation of fine carbides and the reduction in compressive residual stress [15]. Compared with studies on ferrous metals, research on the effect of DCT on nonferrous metals such as Mg, Al, and Ti alloys are limited. Asl et al. [2] found that after DCT, tiny laminar b phase particles almost dissolve, and the coarse divorced eutectic b phase extends into the neighboring matrix. As a result, the mechanical properties of AZ91 Mg alloy significantly improve. Jiang et al. [16] found that DCT induces the refinement of grains to approximately 0.1–3.0 lm, thereby improving the strength and elongation of 3102 Al alloy foil. As aforementioned, evident changes in microstructure such as phase transformation or grain refinement have been verified for DCT-treated alloys. However, changes in microstructure and mechanical properties that can be introduced by DCT if no phase transformation occurs are worth investigating. In this study, coarse-grained zirconium (Zr) with a hexagonal close packed (hcp) structure and higher microstructure stability at cryogenic
514
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
temperature was chosen as a model metal because of the limited deformation capability that results in easily saturated dislocation density [17]. Some unusual changes in mechanical properties were found in DCT-treated Zr, and the possible mechanisms in terms of microstructural modification were discussed. 2. Experimental The material used in this study was commercial pure Zr plate (chemical composition summarized in Table 1). The plate was annealed in a vacuum at 1123 K for 4 h to obtain a homogeneous polycrystalline structure. The average grain size of the as-annealed sample was approximately 30–50 lm. The samples cut from the as-annealed plate were placed in a liquid nitrogen environment at 77 K for 24 h. Afterwards, the samples were taken out and cooled to room temperature. X-ray diffraction (XRD) patterns of the as-annealed and DCT-treated samples were measured using a Rigaku D/max 2500 X-ray diffractometer (18 kW) with Cu Ka radiation (wavelength k = 1.54 Å) in continuous-scanning mode over 2h = 30– 80° at a step of 0.02°. Differential scanning calorimetry (DSC) was conducted on a Diamond DSC (PerkinElmer Inc., UK). For dynamic scans, samples were subjected to temperatures ranging from 20 °C to 160 °C at a cooling rate of 20 °C/min. Microstructure features before and after DCT were observed by optical microscopy (OM), electron backscatter diffraction (EBSD), and transmission electron microscopy (TEM). EBSD analysis was performed using a HITACHI S-4800 scanning electron microscopy (SEM) system, whereas TEM observations were carried out with a JEOL-2010 TEM system at a voltage of 200 kV. A FM-ARS9000 Vickers microhardness tester was used to measure the hardness of samples. Indentations were carried out with a given load of 200 g and a dwell time of 10 s. Hardness distributions over a square area of 4.5 4.5 mm2, with a space between adjacent indentations of 0.5 mm were obtained. The tensile samples were machined with a gauge length of 30 mm and a cross-section of 10 1.5 mm2. Four repeated tensile tests were performed on a MTS test system at a strain rate of 1 104 s1 at room temperature under extensometer-measured strain control. Fracture surfaces after the tensile test were observed by SEM, and microstructures near the fracture surfaces were observed by TEM.
3. Results and discussion 3.1. XRD and DSC analyses Fig. 1(a) shows the XRD patterns of as-annealed and DCTtreated samples. No phase transformation was observed in DCTtreated Zr compared with as-annealed Zr. However, the intensities of the diffraction peaks of DCT-treated Zr were slightly enhanced, which can be attributed to the variation in lattice constants caused by DCT. Changes in lattice constants were also observed in DCTtreated Mg [18,19]. To further confirm the microstructure stability of Zr at cryogenic temperature, a cryogenic DSC curve was obtained and is shown in Fig. 1(b). The smooth feature of the DSC curve indicated that no transformation occurred during DCT. 3.2. EBSD analysis To gain deep insight into microstructural changes, EBSD analysis was performed. In this analysis, orientation imaging microscopy (OIM) maps, inverse pole figure (IPF) maps, and boundary misorientation angle distributions were obtained before and after DCT, as shown in Fig. 2. Fig. 2(a) and (b) shows the OIM maps of Zr before and after DCT, with high-angle (HAGBs; grain boundary misorientations P 15°) and low-angle (LAGBs; grain boundary misorientations < 15°) grain boundaries depicted by black and white lines, respectively. The crystallographic directions corresponding to various colors can be inferred from the IPF triangle shown at the bottom right corner of Fig. 2(a) and (b). The microstructure was characterized by equiaxed grains and a higher
Fig. 1. XRD patterns of as-annealed and DCT-treated samples (a), and the cryogenic DSC curve of Zr (b).
fraction of HAGBs. No phase transformation and precipitation occurred during DCT, consistent with the XRD and DSC results. Fig. 2(c) and (d) shows the IPF maps of Zr before and after DCT, respectively. The maximum intensity indicated the number of random orientations. A strong preference was observed toward the (0 0 0 1) orientation, which was corroborated by the OIM maps in Fig. 2(a) and (b). Furthermore, the intensity of prism planes was lower and the grain orientations were much closer to the (0 0 0 1) basal plane orientation after DCT. This finding can be attributed to the ordered arrangement of atoms resulting from the small changes in lattice constants [18,19]. The misorientation angle distributions in Fig. 2(e) and (f) shows that the crystallite boundaries were mainly high angle in nature. The HAGB fractions of Zr before and after DCT were about 81.2% and 85.3% of the total grain boundary length, respectively. Thus, DCT improved the formation of HAGBs. The increase in HAGB fractions may be related to the extrinsic dislocations formed during DCT (Fig. 3). The extrinsic dislocations were probably easier to react with the dislocations in LAGBs [20,21]. Therefore, the accumulation and rearrangement of dislocations resulted in the transformation of LAGBs to HAGBs, leading to increased HAGB and decreased LAGB fractions. 3.3. TEM observations
Table 1 Chemical composition (in wt%) of the investigated pure Zr. Fe + Cr
C
N
H
O
Hf
Zr
0.2
0.05
0.01
0.005
0.16
4.5
Balance
The TEM images of the samples before and after DCT are presented in Fig. 3. The dislocation density of the DCT-treated sample was apparently higher than that of the as-annealed sample. When the temperature dropped from room temperature to cryogenic
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
515
Fig. 2. Representative OIM maps of (a) as-annealed and (b) DCT-treated samples. IPF maps of (c) as-annealed and (d) DCT-treated samples. Distribution of boundary misorientation angle of (e) as-annealed and (f) DCT-treated samples.
temperature, the volume of the material contracted. This volume contraction released great compression deformation energy that served as the driving force for the formation and movement of dislocations [18]. The improvement in dislocation density played a significant role in enhancing mechanical properties and deformation behavior, as discussed in the following sections. 3.4. Hardness measurements To better understand the variation in hardness of Zr, contour maps of the spatial distribution of the Vickers hardness (Hv) over a 4.5 4.5 mm2 square area before and after DCT are shown in Fig. 4(a) and (b), respectively. The hardness of as-annealed Zr
substantially varied from approximately 140–190 Hv, with an average value of 159.5 Hv. By contrast, the hardness of DCT-treated Zr generally changed from approximately 160–220 Hv, with an average value of 197.5 Hv. Evidently, the average hardness significantly increased after DCT. The difference between the two average hardness values reached 23.8%. Although the spatial distribution of hardness was not uniform from one area to another, the increase in hardness of the DCT-treated sample was a holistic rather than localized phenomenon, as can also be confirmed by the hardness evolution characterized by the color maps in Fig. 4. The inhomogeneity of hardness distribution may be closely related to the grain orientation. An in situ comparison between the OIM map and OM image of hardness indentation of DCT-treated Zr is presented
516
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
Fig. 3. TEM images of (a) as-annealed and (b) DCT-treated samples.
among different grain orientations reached up to 30.2%. The dependency of hardness values on orientation has also been observed in previous reports [22–24]. For hcp Zr, the axial ratio c/a was 1.593, which was less than that of the ideal value (i.e., <1.633). Consequently, the lattice resistance for prism planes was lower than that for basal or pyramidal planes; thus, slip preferred to occur on prism planes [25]. In other words, plastic deformation more easily occurred on prism planes than on basal planes. Consequently, the hardness in basal planes was much larger than that in prism planes. Therefore, the overall improvement of hardness may be closely related to both changes in grain orientation and dislocation density. After DCT, the grain orientation tended to be more consistent, resulting in ordered strengthening [18]. Meanwhile, the increase in dislocation density increased the force to overcome the resistance of dislocation motion. Thus, the improvement in plastic deformation resistance resulted in increased hardness. 3.5. Tensile properties The tensile engineering and true stress–strain curves of as-annealed and DCT-treated Zr are presented in Fig. 6. For as-annealed Zr, the yield strength (0.2% offset, r0.2) and ultimate tensile strength (rUTS) were 288 and 393 MPa, respectively. The elongation to fracture reached 28.7%. Correspondingly, r0.2 and rUTS of DCT-treated Zr increased by about 60% and 40%, respectively, compared with as-annealed Zr. Meanwhile, the elongation to fracture reached 26.5%, which was similar to that of as-annealed Zr. This finding indicated that DCT was a better route to improving the strength and keeping a good ductility of Zr. As depicted in Fig. 6(b), DCT-treated Zr also exhibited high strain hardening capabilities relative to as-annealed Zr. The strain hardening exponent (n) is defined by [26]
r ¼ K en
Fig. 4. Spatial distributions of the hardness values of Zr: (a) before and (b) after DCT.
in Fig. 5. The letters in Fig. 5(a) correspond to those in Fig. 5(b), and grain M indicates the reference grain. The hardness values of grains A–D in Fig. 5(b) were 211.65, 259.54, 199.35, and 233.57 Hv, respectively. The hardness values varied with the orientation between basal and prism planes. Planes close to the (0 0 0 1) basal plane orientation were significantly harder than planes close to 0Þ and ð2 1 1 0Þ planes. The hardness difference the prism ð1 0 1
ð1Þ
where r is the true stress, e is the true strain, and K is a constant. The n values for as-annealed and DCT-treated Zr were 0.09 and 0.11, respectively. The enhancement in strain hardening capabilities of DCT-treated Zr was further confirmed by an improved strain hardening rate (H) defined by [26]
H¼
1 @r r @e
ð2Þ
Variations in H with e before and after DCT are displayed in Fig. 7. The variation trends were consistent with continuous strain hardening to significant strains, but the strain hardening rate of the DCT-treated sample was larger than that of the as-annealed sample.
517
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
(a)
M
A
(b)
D B
C
B
C D A
M
Fig. 5. (a) OIM map and (b) corresponding OM image of hardness indentation of the DCT-treated Zr.
Fig. 7. Normalized strain hardening rate (H) against the true strain of as-annealed and DCT-treated samples.
Fig. 6. Tensile (a) engineering and (b) true stress–strain curves of as-annealed and DCT-treated samples.
DCT-treated Zr exhibited enhancement of strength and maintained the good ductility compared with as-annealed Zr. The variation in mechanical properties can be attributed to the difference in dislocation density and dislocation configurations produced during deformation [17,27,28]. To reveal the mechanism underlying the increase in strength and maintain the good ductility in DCTtreated Zr, TEM observations of the fractured tensile samples were performed (Fig. 8). As shown in Fig. 8(a), for as-annealed Zr after tensile deformation, the grains were apparently distorted and elongated, and masses of dislocations were found in the grains. However, the deformation of DCT-treated Zr was more uniform and
compatible than that of as-annealed Zr. Many dislocations were also observed within the grains, which were divided into smaller blocks (Fig. 8(b)). During DCT, a large number of pre-existing dislocations were introduced into the grains (Fig. 3(b)), and these grains were movable during tensile deformation [27–32]. At the initial stage of deformation, more dislocations and dislocation tangles formed. The increase in dislocation density and resistance of dislocation motion resulted in improved strength. With increased strain, many irregular dislocation tangles changed into parallel dislocation lamellas along with the tensile direction. With further increased strain, dislocation accumulation and rearrangement occurred to form dislocation cells, dense dislocation walls, and nanoscale subgrains and grains. As a result, the coarse grains were divided into smaller blocks. These blocks made the plastic deformation of DCT-treated Zr more uniform and compatible, thereby resulting in the ductility being comparable to that of as-annealed Zr. 3.6. Fracture behavior The fracture surfaces of the as-annealed and DCT-treated Zr sample are shown in Fig. 9. As shown in Fig. 9(a), the fracture surface of edge regions of the as-annealed sample exhibited mixed characteristics with dimples and a small amount of quasi-cleavage facets, whereas the fracture surface in the middle region as shown in Fig. 9(b) mainly contained many quasi-cleavage facets. The dimples in the edge of fracture surface of as-annealed sample were related to the noticeable necking caused by severe strain
518
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
Fig. 8. TEM images of (a) as-annealed and (b) DCT-treated samples after tensile tests.
(a)
(b)
500 μm
500 μm
(c)
(d)
500 μm
500 μm
Fig. 9. SEM images showing the fracture surfaces of as-annealed and DCT-treated Zr tensile samples. (a) and (b) are the edge and middle region of as-annealed Zr sample; (c) and (d) are the edge and middle region of DCT-treated Zr sample.
localization during tensile test. For comparison, the fracture surfaces of DCT-treated Zr in the same regions are presented in Fig. 9(c) and (d), respectively. The fracture characteristics of two regions were mainly consisted of quasi-cleavage facets. No obvious necking was observed in the edge and middle regions of the fracture surfaces. The above findings indicated that DCT restricted strain localization occurring in the edge regions of tensile samples. For asannealed Zr, plastic instability (i.e., necking) occurred in the edge region of tensile samples before ultimate fracture. Moreover, Fig. 2(e) and (f) shows that the fraction of LAGBs before DCT was slightly larger than that after DCT. LAGBs are reportedly associated with the early onset of plastic instability during tensile testing [33–35]. The plastic instability accounted for the different fracture
characteristics in the edge and middle region exhibited by asannealed Zr. By contrast, the motion of pre-existing dislocations and specific dislocation configurations effectively enhanced the compatible deformation capability, resulting in the same fracture characteristics in both regions of the DCT-treated sample. 4. Conclusion Based on the above-described investigations, the following conclusions were drawn. (1) The grain orientations were much closer to the (0 0 0 1) basal plane orientation after DCT. DCT increased HAGB fractions and dislocation density.
C. Yuan et al. / Journal of Alloys and Compounds 619 (2015) 513–519
(2) Hardness was closely related to grain orientations and dislocation density. The hardness in basal planes was much higher than that in prism planes. After DCT, dislocation density increased the resistance of dislocation motion, thereby improving hardness. (3) Strength levels were high and good ductility could still be achieved after DCT. The enhancement in strength was due to the increase in dislocation density, and the maintenance of good ductility resulted from the motion of pre-existing dislocations and specific dislocation configurations. (4) The fracture surfaces exhibited mixed features with quasicleavage facets and dimples for as-annealed Zr and quasicleavage facets for DCT-treated Zr. The grains were divided into smaller blocks because of the interaction of dislocations during the tensile process of DCT-treated Zr. Consequently, the compatible deformation capability of materials was enhanced and good ductility could still be achieved.
Acknowledgment Financial support from the National Basic Research Program of China (Grant No. 2010CB731606) is greatly acknowledged. References [1] [2] [3] [4]
R.F. Barron, Cryogenics 22 (1982) 409–414. K.M. Asl, A. Tari, F. Khomamizadeh, Mater. Sci. Eng., A 523 (2009) 27–31. M. Koneshlou, K.M. Asl, F. Khomamizadeh, Cryogenics 51 (2011) 55–61. A. Molinari, M. Pellizzari, S. Gialanella, G. Straffelini, K.H. Stiasny, J. Mater. Process. Technol. 118 (2001) 350–355. [5] P. Baldissera, Mater. Des. 31 (2010) 4725–4730. [6] P. Baldissera, C. Delprete, Open Mech. Eng. J. 2 (2008) 1–11. [7] B. Podgornik, F. Majdic, V. Leskovsek, J. Vizintin, Wear 288 (2012) 88–93.
519
[8] R. Thornton, T. Slatter, A.H. Jones, R. Lewis, Wear 271 (2011) 2386–2395. [9] T. Slatter, R. Lewis, A.H. Jones, Wear 271 (2011) 1481–1489. [10] P.J. Singh, S.L. Mannan, T. Jayakumar, D.R.G. Achar, Eng. Fail. Anal. 12 (2005) 263–271. [11] S. Zhirafar, A. Rezaeian, M. Pugha, J. Mater. Process. Technol. 186 (2007) 298– 303. [12] V. Leskovšek, B. Podgornik, Mater. Sci. Eng., A 531 (2012) 119–129. [13] H.H. Liu, J. Wang, B.L. Shen, H.S. Yang, S.J. Gao, S.J. Huang, Mater. Des. 28 (2007) 1059–1064. [14] P.F. Stratton, Mater. Sci. Eng., A 449 (2007) 809–812. [15] A. Bensely, S. Venkatesh, D. Mohan Lal, G. Nagarajan, A. Rajadurai, Krzysztof Junik, Mater. Sci. Eng., A 479 (2008) 229–235. [16] X.Q. Jiang, N. Li, H. He, X.J. Zhang, C.C. Li, H. Yang, Mater. Sci. Forum 546–549 (2007) 845–848. [17] D.F. Guo, M. Li, Y.D. Shi, T.Y. Ma, Z.B. Zhang, T.S. Wang, X.Y. Zhang, Mater. Sci. Eng., A 581 (2013) 133–139. [18] J.W. Liu, G.F. Li, D. Chen, Z.H. Chen, Chin. J. Aeronaut. 25 (2012) 931–936. [19] X.F. Zhang, G.H. Wu, W.C. Liu, W.J. Ding, Trans. Nonferrous Met. Soc. China 22 (2012) 2883–2890. [20] P. Xue, B.L. Xiao, Z.Y. Ma, Mater. Sci. Eng., A 532 (2012) 106–110. [21] Y.H. Zhao, J.F. Bingert, Y.T. Zhu, X.Z. Liao, R.Z. Valiev, Z. Horita, T.G. Langdon, Y.Z. Zhou, E.J. Lavernia, Appl. Phys. Lett. 92 (2008) 081903. [22] D.N. French, D.A. Thomas, Trans. AIME 233 (1965) 950–952. [23] T. Takahashi, E.J. Freise, Philos. Mag. 12 (1964) 1–8. [24] B. Roebuck, P. Klose, K.P. Mingard, Acta Mater. 60 (2012) 6131–6143. [25] L. Zhang, Y. Han, J. Lu, Nanotechnology 19 (2008) 165706. [26] G.E. Dieter, Mech. Metall., third ed., McGraw-Hill, London, 1988. [27] D.F. Guo, M. Li, Y.D. Shi, Z.B. Zhang, T.Y. Ma, H.T. Zhang, X.Y. Zhang, Mater. Sci. Eng., A 558 (2012) 611–615. [28] D.F. Guo, M. Li, Y.D. Shi, Z.B. Zhang, H.T. Zhang, X.M. Liu, X.Y. Zhang, Mater. Lett. 66 (2012) 305–307. [29] X.X. Huang, Scr. Mater. 60 (2009) 1078–1082. [30] L. Lu, X. Chen, X. Huang, K. Lu, Science 323 (2009) 607–610. [31] X.X. Huang, N. Hansen, N. Tsuji, Science 312 (2006) 249–251. [32] Z.W. Shan, Raja K. Mishra, S.A. Syed Asif, Oden L. Warren, Andrew M. Minor, Nat. Mater. 7 (2008) 115–119. [33] R. Kapoor, N. Kumar, R.S. Mishra, C.S. Huskamp, K.K. Sankaran, Mater. Sci. Eng., A 527 (2010) 5246–5254. [34] P.L. Sun, C.Y. Yu, P.W. Kao, C.P. Chang, Scr. Mater. 52 (2005) 265–269. [35] N. Kumar, R.S. Mishra, C.S. Huskamp, K.K. Sankaran, Mater. Sci. Eng., A 528 (2011) 5883–5887.