- Email: [email protected]
Tougher TiAl alloy via integration of hot isostatic pressing and heat treatment
Materials Science & Engineering A 688 (2017) 371–377
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Tougher TiAl alloy via integration of hot isostatic pressing and heat treatment
MARK
⁎
Liu Chen , Langping Zhu, Yongjun Guan, Bao Zhang, Jiancong Li Beijing Institute of Aeronautical Materials, Beijing, China
A R T I C L E I N F O
A BS T RAC T
Keywords: Mechanical properties Hot isostatic pressing Intermetallics Titanium aluminide
A special processing method integrating the hot isostatic pressing and heat treatment was developed to produce TiAl alloy. As compared with the traditional two-step method, which consists of separate hot isostatic pressing and subsequent heat treatment in vacuum, the integrated approach will introduce an isostatic pressure throughout the whole process from the consolidation of powder to the heat treatment of coupon. Two types of microstructure, i.e., lamellar and duplex, have been generated by integrated and separate methods. Tensile test at room temperature indicates that the yield strength and tensile elongation of lamellar microstructure generated by integrated method are 650 ± 30 MPa and 2.1 ± 0.2% respectively, which are improved simultaneously as compared with 550 ± 28 MPa and 0.6 ± 0.1% for sample generated by separate approach. Moreover, such simultaneous enhancement of yield strength and ductility is also observed in duplex microstructure, where the yield strength and tensile elongation increase from 420 ± 18 MPa and 2.4 ± 0.4% in samples with separate method, to 540 ± 25 MPa and 3 ± 0.3% in those generated by integrated approach. Additionally, the microstructural examination also revealed the influence of microstructure to the mechanical performance. The results show that the simultaneous improvement of yield strength and tensile elongation is mainly attributed to the suppression of cracking, which is prone to happen during heat treatment without atmospheric pressure. Using the integrated method, the isostatic pressure could sustain the equilibrium pores during heat treatment, and provides an exterior force to balance the potential internal stresses due to phase transformation.
1. Introduction
components, an elongation to failure of 3% would be able to meet the requirements in industry [13]. The optimization of mechanical properties for TiAl alloys could be achieved through micro-alloying and microstructure controlling, which have produced attractive synergies of tensile strength and ductility [14–19]. The mechanical properties of TiAl intermetallic alloys are between brittle ceramics and ductile metals. When deformed at room temperature, a little dislocation behavior could be activated in equiaxed γ grains or special oriented γ lamellae, which results in strain hardening and thus uniform elongation [20,21]. However, the mechanical properties of intermetallic TiAl alloys are extremely sensitive to defects in microstructure. Therefore, the defects such as microcrack in TiAl alloys always result in premature failure during tensile deformation, leading to both low tensile strength and low elongation to failure [22]. In addition, in order to obtain an optimized combination of strength, ductility and toughness, heat treatments are always required to adjust the microstructure of TiAl alloys. Since the solid phase transformations are always prevalent and there is no atmospheric pressure during heat treatment, those microcracks would be generated as a result of internal
TiAl intermetallic alloy is one of the most potential candidates for advanced aero-engine materials due to its high specific strength at elevated temperatures [1–6]. However, the extremely low tensile ductility at room temperature strongly hinders its applications in industry. For instance, the elongation to failure of many TiAl alloys is typically less than 1% at room temperature, which results in low damage tolerance and makes it difficult for manufacturing and component assembling. Therefore, strategies to improve the mechanical properties of TiAl alloys at ambient temperature have attracted much interest during last two decades [7–12]. In contrast to ductile metals and alloys, a significant improvement of tensile ductility for TiAl intermetallic alloys at room temperature is practically impossible and dispensable. The lattice dislocations of intermetallic alloys are hardly to be activated at relatively low temperature. Once they were activated substantially (followed by high tensile ductility in logic), their high temperature strength would be deteriorated drastically. Based on the design criteria for aero-engine
⁎
Corresponding author. E-mail address: [email protected] (L. Chen).
http://dx.doi.org/10.1016/j.msea.2017.02.028 Received 15 December 2016; Received in revised form 6 February 2017; Accepted 7 February 2017 Available online 08 February 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
the cans filled with TiAl powders were compacted via HIP technique at 1533 K and 160 MPa for 4 h with heating rate of about 4 K per min, and then furnace cooled to room temperature. The stainless steel containers were removed by lathe machining and consolidated bulk samples were obtained (designated as HIPed sample). In order to generate duplex or lamellar structure, the HIPed samples were annealed at 1553 K for 4 h or 1633 K for 0.5 h in vacuum, with furnace cooling in both conditions (designated as Dsep and Lsep samples, respectively). For IHH processing route, the first HIP stage had the same parameters as that in SHH route. The difference is that at the end of HIP process, temperature were straight elevated to 1553 or 1633 K and maintained for 4 or 0.5 h to generate duplex or lamellar structure, followed by furnace cooling in both conditions (named as Dint and Lint samples, respectively). The gauge size of tested samples is ϕ8 mm × 40 mm , and the tensile tests were carried out using MTS Landmark testing system with uniaxial quasistatic strain rate of 2 × 10−4 s−1 at room temperature. The microstructure was investigated by backscattered electronic (BSE) imaging using a field emission gun scanning electron microscope. Specimens for BSE investigation were electro-polished using a solution of 95% ethylalcohol and 5% perchloric acid (HClO4) at 253 K with voltage of 40 V.
stresses, which make the sample or component defective prior to testing or service. Accordingly, the modifications in traditional heat treatment processes to suppress the formation of defects and produce the defect-free samples or components would be significant step towards enhancing mechanical properties in TiAl alloys. Atmospheric pressure during heat treatment has been developed in special field to recover the mechanical properties of turbine blades in industry [23,24]. With a hot isostatic pressing atmosphere, the microvoids or microcracks could be reclosed and thus the rejuvenated mechanical properties would be obtained. Inspired by the advantages of pressure to heat treatment, we have taken the hot isostatic pressing (HIP) technique to provide a pressure environment during heat treatment to suppress the potential formation of microcracks in TiAl intermetallic alloys. Based on HIP technique, a special processing method integrating HIP and heat treatment (IHH) was designed. For comparison, traditional separated HIP and subsequent heat treatment (SHH) in vacuum was also conducted. The present study explores the economical strategy for improving the mechanical properties of TiAl alloys, especially for components with relatively complex geometries and large-scale sizes. The mechanical properties for samples generated by two different methods were tested and analyzed, and corresponding microstructural evolutions as well as defect-induced deterioration in mechanical properties were investigated in detail.
3. Results 2. Experimental procedures 3.1. Mechanical properties The TiAl alloy used in this study has a nominal chemical composition (at%) of 47Al, 2Cr, 2Nb and Ti in balance. Alloy ingots were first prepared by arc melting and drop casting into Cu molds using commercially pure metals. The ingots were then remelted and atomized in argon atmosphere to generate alloy powders with mean particle size of about 100 µm. Powders were filled into 304 stainless steel cans with diameter of 60 mm and 150 mm in length, followed by degassing at room temperature, 623 K and 873 K for 1, 2, and 4 h, respectively. After degassing, cans were sealed in vacuum of about 10−1 Pa, which will be used for HIP and heat treatment. Two different processing routes were then conducted respectively, which are schematically illustrated in Fig. 1. During SHH processes,
Fig. 2 exhibits the tensile properties of TiAl samples produced by different processing routes. The typical engineering stress‒strain curves are shown in Fig. 2(a) and (b). The mean yield stress and elongation to failure of HIPed samples are measured as 480 ± 21 MPa and 2 ± 0.3%. When subjected to annealing at 1553 K to form duplex microstructure via SHH route, the Dsep samples have an average yield stress and elongation to failure of 420 ± 18 MPa and 2.4 ± 0.4%, while Dint samples produced by IHH route have higher yield stress and elongation to failure up to 540 ± 25 MPa and 3 ± 0.3%. When it comes to Lsep samples with lamellar structure which were annealed at 1633 K, as exhibited in Fig. 2(b), the yield stress and elongation to failure are measured as 550 ± 28 MPa and 0.6 ± 0.1%, respectively. In contrast, the Lint samples treated by IHH route have higher yield stress and elongation to failure up to 650 ± 30 MPa and 2.1 ± 0.2%. According to the results of mechanical properties, the HIPed samples have moderate yield strength and tensile ductility. When the traditional heat treatment in vacuum was applied to generate the lamellar or duplex microstructure, it is hard to achieve a good combination between the strength and tensile ductility. Taking the lamellar structure as an example, the yield stress could be improved; however, such improvement is accompanied by a significant sacrifice of tensile ductility. In contrast, when processed through IHH route, the yield stress and elongation to failure of both duplex and lamellar structures are increased simultaneously. Corresponding to Fig. 2(a) and (b), the true stress and strain hardening rate were plotted against true strain in Fig. 2(c) and (d). As compared with the Dsep sample, the Dint sample produced by IHH route shows higher strain hardening rate during tensile deformation, indicating stronger strain hardening ability and hence enhanced tensile ductility, which is in line with the experimental results exhibited in Fig. 2(a). For the lamellar structure, the Lsep sample is quickly failed at early stage of tensile deformation, while Lint sample sustains to relatively higher tensile strain. According to Considère theory, plastic instability during uniaxial tensile deformation sets in as strain hardening rate decreases to the same level as true stress [25]. However, all samples do not follow the prediction of Considère criterion. Instead, failures all occurred at low strains where the strain hardening rates are still much greater than true stresses, which means that the failure of
Fig. 1. Schematic illustration of two different processing routes. (a) Traditionally separated route (SHH) HIPing first at 1533 K and 160 MPa for 4 h, followed by heat treatment in vacuum at 1633 and 1553 K for 0.5 and 4 h, respectively. (b) Integrated processing route (IHH) undertaking heat treatment immediately at the end of HIP with pressure atmosphere of 160 MPa.
372
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 2. Mechanical properties of TiAl alloy. (a) Engineering stress‒strain curves for duplex microstructure generated by IHH (Dint) and SHH (Dsep). The curve of HIPed sample was also plotted for comparison. (b) Engineering stress‒strain curves for lamellar microstructure generated by IHH (Lint) and SHH (Lsep) routes. Corresponding true stress (σt) and strain hardening rate (Θ) plotted against the true strain for (c) duplex and (d) lamellar structure, respectively.
to the gray contrast of micrograph, some Al atoms have segregated in γ grains, as denoted by the yellow asterisks [27], while some heavy atoms (e.g. Nb or Cr) have segregated in lamellar regions, which are denoted by the red asterisks. The strong atom segregation implies that the HIPed sample is far from equilibrium state, which is one of the reasons to accelerate the solid state phase transformation during heat treatment stage. Fig. 4 presents the micrographs of duplex structure generated by two different processing routes. Fig. 4(a) shows the duplex structure of Dsep sample, which consists of nearly equiaxed γ grains and lamellar α2/γ domains in about half and half proportion. Fig. 4(b) is the magnified micrograph corresponding to encircled area in Fig. 4(a). Some microcracks are indicated by white arrows in Fig. 4(a) and (b), with length ranging from several 10–100 s of microns. In contrast to Dsep samples, Dint samples produced by IHH route exhibit duplex structure free of microcrack. Fig. 4(c) shows a typical microstructure of Dint sample, and the area marked by rectangular line is enlarged and shown in Fig. 4(d). The area fractions of lamellar colonies were counted as about 41% and 43% in Dsep and Dint samples. In order to verify that the Dint sample is free of microcracks, more than 70% cross-sectional area had been screened, and no visible cracks were detected during screening. Based on BSE micrographs, the sizes of different microstructural characteristics were counted and their statistical distributions are shown in Fig. 5. For Dsep sample produced by SHH route, Fig. 5(a)– (c) display the size distributions of γ grain, α2/γ domain and α2/γ lamellar spacing, and their average sizes are counted as 55.7, 61.9 and 2.5 µm, respectively. For comparison, corresponding size distributions of Dint sample produced by IHH route are displayed in Fig. 5(d)–(f), and their mean size are 43.7, 69.6 and 2.3 µm. The inset in Fig. 5(c) illustrates schematically the definition of lamellar spacing, which equals to the sum of mean thickness of α2 and γ lamellae. According to statistical results of characteristic size in lamellar structure, one can conclude that the isostatic pressure during heat treatment stage of IHH processing has little influence on the overall
TiAl alloy is attributed to the premature fracture, rather than its intrinsic poor mechanical properties [26]. Hence, approaches to suppress the premature failure would be beneficial to improve the mechanical properties of TiAl alloy.
3.2. Microstructure observation BSE micrographs revealed the influence of processing routes on the evolution of microstructures for TiAl alloy. Fig. 3 shows the micrograph of HIPed sample, which is extremely heterogeneous in microstructure. The regions of equiaxed γ grains account for about 45% in area fraction, and the rest are mainly fine stripe-like colonies. According
Fig. 3. BSE micrograph of HIPed sample consisting of equiaxed γ-TiAl grains and stripelike colonies. As denoted by asterisks (yellow and red ones correspond to Al- and Ti-rich regions, respectively), atom segregations are shown in both equiaxed and stripe-like regions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
373
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 4. BSE micrographs of duplex structure generated by SHH and IHH routes, consisting of lamellar colonies and equiaxed γ grains. (a) Micrograph correlated to SHH route showing microcracks as marked by arrows. (b) Corresponding high magnification image of the enclosed area in (a). (c) Micrograph without cracks for samples produced by IHH route. (d) Enlarged view of corresponding enclosed area in 4(c) indicating defect-free duplex structure. (α2 lamellae in lamellar colony is distinguished by brighter contrast due to its higher concentration of Ti).
Fig. 5. Statistical distribution of different microstructural sizes for duplex structure. Size distributions of (a) γ grain, (b) α2/γ domain and (c) lamellar spacing for Dsep sample produced by SHH route. Corresponding size distributions of (d) γ grain, (e) α2/γ domain and (f) lamellar spacing for Dint sample produced by IHH route.
374
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 6. BSE micrographs of lamellar structure generated by SHH and IHH routes. (a) Micrograph correlated to SHH route showing microcracks (marked by white arrows). (b) High magnification image corresponding to enclosed area of (a) exhibiting defective lamellar structure. (c) Micrograph free of cracks for samples generated by IHH route. (d) Enlarged view of corresponding enclosed area in (c).
Fig. 7. Statistical distribution of different microstructural sizes for lamellar TiAl. Size distributions of (a) α2/γ domain and (b) lamellar spacing for lamellar structure correlated to SHH route. Corresponding size distributions of (c) α2/γ domain and (d) lamellar spacing for lamellar structure generated by IHH route.
375
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
treatment without exterior pressure would break the equilibrium condition and thus possibly result in swelling in microstructure [30]. For TiAl alloys produced by powder metallurgical methods, the subsequent heat treatment is always necessary and important to adjust the microstructure for optimized lamellar or duplex morphology. Therefore, the swelling tendency during heat treatment stages without atmospheric pressure will increase the possibility of debonding and cracking. Additionally, the heat treatment process involves substantial solid phase transformation for TiAl alloys, during which the displacive transformation [31] could result in both microscopic and macroscopic stresses [32]. Accordingly, during the process of heat treatment, under the internal stress resulting from the breaking of equilibrium in pores and the solid phase transformation, the weak-bonding area as a result of load shielding would be failed and then the cracking will occur. Based on the microstructure observations, both lamellar spacing and phase fractions are not obviously changed due to isostatic pressure at heat treatment stages in either duplex (Fig. 5) or lamellar (Fig. 7) microstructures. Therefore, in present investigation, the isostatic pressure during heat treatment will mainly sustain the equilibrium in pores, and provide an exterior force to balance the potential internal stresses due to phase transformation. As a result, the potential cracking was suppressed and the crack-free microstructure was obtained finally in TiAl alloy.
evolution of microstructure. As shown in Fig. 5, the Dint sample has similar microstructural sizes in terms of γ grain, lamellar colony and lamellar spacing, as those in Dsep sample. However, it should be noted the area fraction of lamellar colonies is slightly improved in Dint sample, and the lamellar spacing size in Dint sample is a little finer than that in Dsep sample. It seems that the pressure during IHH processing has a slight positive effect on the formation of lamellar structure. Fig. 6 shows the micrographs of lamellar structure generated by different processing routes. As shown by Fig. 6(a), some microcracks are indicated by white arrows, and the enclosed area is further displayed as enlarged view in Fig. 6(b). The microcracks indicate that Lsep samples which were produced by SHH route are defective in microstructure prior to tensile deformation. However, as shown by Fig. 6(c) and (d), the Lint sample produced by IHH route shows lamellar structure free of microcracks, indicating successful suppression of crack formation due to isostatic pressure. Additionally, more than 70% cross-sectional area of Lint sample had been screened and no visible cracks were detected. Fig. 7 shows statistical distributions of lamellar domain size and lamellar spacing for Lsep and Lint samples. Fig. 7(a) and (b) display the distribution of domain size and lamellar spacing of Lsep sample, while corresponding distributions of Lint sample are depicted in Fig. 7(c) and (d). The mean lamellar domain sizes in Lsep and Lint sample are close to each other, which equal to 75.4 and 70.4 µm, respectively. Similar to the size of lamellar domain, the average lamellar spacing between Lsep and Lint sample has no obvious difference in magnitude. According to Figs. 6 and 7, it could be concluded that isostatic pressure during IHH route has little influence on the overall evolution of microstructure, but strongly suppressed the formation of microcracks in lamellar structure. Therefore, the optimized mechanical properties of samples produced by IHH route vis-à-vis those processed by SHH route are mainly attributed to their crack-free microstructures.
4.2. Influence of microcrack on mechanical properties The mechanical properties of intermetallic alloys are extremely sensitive to defects in microstructure. With regards to TiAl alloys which have relatively high brittle-to-ductile transition temperatures [33,34], microcracks are extremely harmful in deteriorating mechanical performance at room or even at low temperatures. Generally, microcrack will decrease the tensile elongations of materials with low ductility, since their strain hardening will be suppressed and the samples are mostly elongated via crack propagation rather than traditional plastic straining. That is the reason why many intermetallics like TiAl alloy always exhibit negligible tensile elongation. However, the influence of initial defects such as pore or microcrack to the yield strength seems not quite as much as that to the tensile elongation, which is anomalous to the current results where yield strength is also decreased due to microcracks (Fig. 2(a) and (b)). For example, the commercially pure Ti with different porosities [35] and the Al-Si alloy with different intensity of damage [36] have shown significant reduction of tensile elongation but little change of yield strength with increase of defects. For current TiAl alloy, however, both yield strength and tensile elongation will be decreased if microcracks exist within the microstructure. The primary reason for the difference between present study and literatures [35,36] might be that such materials are more ductile, in which the propagation of defects will be initiated after the dislocationdominated yielding stage. However, during tensile deformation at room temperature for TiAl alloy, the nucleation of cracks had been detected even prior to macroscopic yielding by acoustic emission technique [22,37], manifesting that the cracking behavior had happened along with the dislocation gliding during yielding process. The microcracks can be stable at least within the size comparable to the colony size or grain diameter [10], and thus the tensile process will involve the dislocation and cracking behavior at the same time. Therefore, the yield strength is not only determined by the activation of dislocation gliding, but also the morphology and geometry of microcracks. The microcrack will directly reduce the effective crosssectional area to bear the applied load, leading to decrease of yield strength. In addition, similar to the deformation in composites, the propagation of crack could produce a strain besides the traditional elastic and plastic strains, lowering down the flow stress curves and showing lower apparent yield stresses. Since the brittle materials are more sensitive to crack behavior in mechanical property, both their yield strength and tensile elongation will be deteriorated due to
4. Discussion 4.1. Benefits of isostatic pressure to crack-free microstructure During hot isostatic pressing, the metallic powders inside the can will deform, diffuse and finally be consolidated into the desired solid geometry with the thermal and dynamic assistance of temperature and pressure. However, it is still difficult in reality to remove the artifacts from final bulk samples or components, which is therefore harmful to their mechanical performance, especially for materials with low ductility. Load shielding is one of the possible reasons resulting in defects in microstructure. As proposed by Wadley et al. in their investigation, can shielding will decrease the pressure supported by powder, making the real pressure exerted on powder less than that exerted on can, especially at relatively low temperature or when the can has a yield strength comparable to or greater than that of the powder [28]. In addition, for cylindrical can with greater size in length than that in diameter, the load shielding is more severe along the axis direction than that along the radius direction [29], indicating the size dependence of load shielding in magnitude. Therefore, the bigger the sample or component is, the stronger the load shielding will become. For present investigation, the 304 stainless steel was chose as the can material, which will lower down the can shielding effect due to its low flow stress at HIP temperature. However, the relatively large size in diameter and especially in length would aggravate the load shielding, which decreases the real pressure exerted on powder and thus increase the probability to produce pores and weak bonding. Furthermore, since the can interior is not in absolute vacuum condition, the pores at final stage of HIP will be gas-filled due to the concentration of gas molecules, and the equilibrium pressure inside will then make such pores difficult to be closed. Even if the pores would be closed finally by HIP and give a full density, the subsequent heat 376
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
References
microcrack. The present result is in alignment with the investigation in low-ductile Ti-6Al-4V alloys with different porosities [35], where both yield strength and tensile elongation were decreased with the increase of defects. Accordingly, the microcracks should be the primary reason for the low yield strength and tensile ductility for duplex and lamellar samples produced by SHH route. Therefore, special integrated processing route with isostatic pressure during entire process has benefits to suppress the formation of microcrack in both duplex and lamellar microstructures, and provides improved combination between yield strength and tensile elongation.
[1] F. Appel, H. Clemens, F.D. Fischer, Prog. Mater. Sci. 81 (2016) 55–124. [2] F. Appel, J.D.H. Paul, M. Oehring, Gamma Titanium Aluminide Alloys: Science and Technology, John Wiley & Sons, Weinheim, 2011. [3] Y.W. Kim, JOM 47 (1995) 39–42. [4] Y.W. Kim, D.M. Dimiduk, JOM 43 (1991) 40–47. [5] M. Yamaguchi, H. Inui, K. Ito, Acta Mater. 48 (2000) 307–322. [6] W. Wallgram, T. Schmölzer, L. Cha, G. Das, V. Güther, H. Clemens, Int. J. Mater. Res. 100 (2009) 1021–1030. [7] C.T. Liu, P.J. Maziasz, J.L. Wright, MRS Proceedings, Cambridge Univ Press, 1996, pp. 83. [8] C.T. Liu, J.H. Schneibel, P.J. Maziasz, J.L. Wright, D.S. Easton, Intermetallics 4 (1996) 429–440. [9] Y. Wu, S.K. Hwang, Acta Mater. 50 (2002) 1479–1493. [10] C.T. Liu, P.J. Maziasz, Intermetallics 6 (1998) 653–661. [11] Y. Song, D.S. Xu, R. Yang, D. Li, Z.Q. Hu, Intermetallics 6 (1998) 157–165. [12] F. Appel, M. Oehring, R. Wagner, Intermetallics 8 (2000) 1283–1312. [13] P. Au, Carleton University, Ottawa, 1997. [14] T. Tetsui, K. Shindo, S. Kobayashi, M. Takeyama, Scr. Mater. 47 (2002) 399–403. [15] Z.C. Liu, J.P. Lin, S.J. Li, G.L. Chen, Intermetallics 10 (2002) 653–659. [16] J.N. Wang, J. Yang, Q. Xia, Y. Wang, Mater. Sci. Eng. A 329–331 (2002) 118–123. [17] E. Schwaighofer, H. Clemens, S. Mayer, J. Lindemann, J. Klose, W. Smarsly, V. Güther, Intermetallics 44 (2014) 128–140. [18] H. Clemens, W. Wallgram, S. Kremmer, V. Güther, A. Otto, A. Bartels, Adv. Eng. Mater. 10 (2008) 707–713. [19] A. Couret, G. Molénat, J. Galy, M. Thomas, Intermetallics 16 (2008) 1134–1141. [20] H.A. Calderon, V. Garibay-Febles, M. Umemoto, M. Yamaguchi, Mater. Sci. Eng. A 329–331 (2002) 196–205. [21] K. Edalati, S. Toh, H. Iwaoka, M. Watanabe, Z. Horita, D. Kashioka, K. Kishida, H. Inui, Scr. Mater. 67 (2012) 814–817. [22] D. Hu, A. Huang, H. Jiang, N. Mota-Solis, X. Wu, Intermetallics 14 (2006) 82–90. [23] A. Baldan, J. Mater. Sci. 26 (1991) 3409–3421. [24] A.G. Peterson, D.W. Gandy, G. Frederick, J.T. Stover, R. Viswanathan, Google Patents, 2002. [25] X.L. Wu, M.X. Yang, F.P. Yuan, G.L. Wu, Y.J. Wei, X.X. Huang, Y.T. Zhu, Proc. Natl. Acad. Sci. USA 112 (2015) 14501–14505. [26] Y.G. Kim, J.M. Han, J.S. Lee, Mater. Sci. Eng. A 114 (1989) 51–59. [27] E. Schwaighofer, B. Rashkova, H. Clemens, A. Stark, S. Mayer, Intermetallics 46 (2014) 173–184. [28] H.N.G. Wadley, R.J. Schaefer, A.H. Kahn, M.F. Ashby, R.B. Clough, Y. Geffen, J.J. Wlassich, Acta Mater. 39 (1991) 979–986. [29] J. Xu, R.M. McMeeking, Int. J. Mech. Sci. 34 (1992) 167–174. [30] H.V. Atkinson, S. Davies, Metall. Mater. Trans. A 31 (2000) 2981–3000. [31] Y. Sun, Philos. Mag. Lett. 78 (1998) 297–305. [32] B. Verlinden, J. Driver, I. Samajdar, R.D. Doherty, Thermo-Mechanical Processing of Metallic Materials, Elsevier, Amsterdam, 2007. [33] C.T. Liu, Scr. Metall. Mater. 27 (1992) 599–603. [34] D.W. Hu, Rare Met. 35 (2016) 1–14. [35] R. Bourcier, D. Koss, R. Smelser, O. Richmond, Acta Mater. 34 (1986) 2443–2453. [36] G. Guiglionda, W. Poole, Mater. Sci. Eng. A 336 (2002) 159–169. [37] R. Botten, X. Wu, D. Hu, M.H. Loretto, Acta Mater. 49 (2001) 1687–1691.
5. Conclusions A special method integrating traditional hot isostatic pressing and heat treatment was developed to improve the mechanical performance of TiAl alloy. The main conclusions are drawn as follows: (1) An integrated processing method has been developed which combines the traditional HIP and heat treatment process. With this method, an isostatic atmospheric pressure is introduced during entire process from consolidation of powder to heat treatment of coupon. (2) According to tensile testing results, both samples with lamellar and duplex microstructure generated by integrated processing route have simultaneous improvement in yield stress and elongation to failure. Compared with samples processed by traditionally separated HIP and heat treatment, the yield stress and elongation to failure of lamellar structure have been improved from 550 ± 28 MPa and 0.6 ± 0.1%, to 650 ± 30 MPa and 2.1 ± 0.2%. While in the duplex structure, they are increased from 420 ± 18 MPa and 2.4 ± 0.4%, to 540 ± 25 MPa and 3 ± 0.3%, respectively. (3) As evidenced by microstructural investigation, the microcracks play an important role in deteriorating mechanical properties of TiAl alloy. By using integrated processing route, both lamellar and duplex structures free of microcrack are generated, and thus an improvement in mechanical properties has been achieved. Acknowledgements The authors would like to thank the financial support by Natural Science Foundation of China with Grant no. 51301187.
377
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Tougher TiAl alloy via integration of hot isostatic pressing and heat treatment
MARK
⁎
Liu Chen , Langping Zhu, Yongjun Guan, Bao Zhang, Jiancong Li Beijing Institute of Aeronautical Materials, Beijing, China
A R T I C L E I N F O
A BS T RAC T
Keywords: Mechanical properties Hot isostatic pressing Intermetallics Titanium aluminide
A special processing method integrating the hot isostatic pressing and heat treatment was developed to produce TiAl alloy. As compared with the traditional two-step method, which consists of separate hot isostatic pressing and subsequent heat treatment in vacuum, the integrated approach will introduce an isostatic pressure throughout the whole process from the consolidation of powder to the heat treatment of coupon. Two types of microstructure, i.e., lamellar and duplex, have been generated by integrated and separate methods. Tensile test at room temperature indicates that the yield strength and tensile elongation of lamellar microstructure generated by integrated method are 650 ± 30 MPa and 2.1 ± 0.2% respectively, which are improved simultaneously as compared with 550 ± 28 MPa and 0.6 ± 0.1% for sample generated by separate approach. Moreover, such simultaneous enhancement of yield strength and ductility is also observed in duplex microstructure, where the yield strength and tensile elongation increase from 420 ± 18 MPa and 2.4 ± 0.4% in samples with separate method, to 540 ± 25 MPa and 3 ± 0.3% in those generated by integrated approach. Additionally, the microstructural examination also revealed the influence of microstructure to the mechanical performance. The results show that the simultaneous improvement of yield strength and tensile elongation is mainly attributed to the suppression of cracking, which is prone to happen during heat treatment without atmospheric pressure. Using the integrated method, the isostatic pressure could sustain the equilibrium pores during heat treatment, and provides an exterior force to balance the potential internal stresses due to phase transformation.
1. Introduction
components, an elongation to failure of 3% would be able to meet the requirements in industry [13]. The optimization of mechanical properties for TiAl alloys could be achieved through micro-alloying and microstructure controlling, which have produced attractive synergies of tensile strength and ductility [14–19]. The mechanical properties of TiAl intermetallic alloys are between brittle ceramics and ductile metals. When deformed at room temperature, a little dislocation behavior could be activated in equiaxed γ grains or special oriented γ lamellae, which results in strain hardening and thus uniform elongation [20,21]. However, the mechanical properties of intermetallic TiAl alloys are extremely sensitive to defects in microstructure. Therefore, the defects such as microcrack in TiAl alloys always result in premature failure during tensile deformation, leading to both low tensile strength and low elongation to failure [22]. In addition, in order to obtain an optimized combination of strength, ductility and toughness, heat treatments are always required to adjust the microstructure of TiAl alloys. Since the solid phase transformations are always prevalent and there is no atmospheric pressure during heat treatment, those microcracks would be generated as a result of internal
TiAl intermetallic alloy is one of the most potential candidates for advanced aero-engine materials due to its high specific strength at elevated temperatures [1–6]. However, the extremely low tensile ductility at room temperature strongly hinders its applications in industry. For instance, the elongation to failure of many TiAl alloys is typically less than 1% at room temperature, which results in low damage tolerance and makes it difficult for manufacturing and component assembling. Therefore, strategies to improve the mechanical properties of TiAl alloys at ambient temperature have attracted much interest during last two decades [7–12]. In contrast to ductile metals and alloys, a significant improvement of tensile ductility for TiAl intermetallic alloys at room temperature is practically impossible and dispensable. The lattice dislocations of intermetallic alloys are hardly to be activated at relatively low temperature. Once they were activated substantially (followed by high tensile ductility in logic), their high temperature strength would be deteriorated drastically. Based on the design criteria for aero-engine
⁎
Corresponding author. E-mail address: [email protected] (L. Chen).
http://dx.doi.org/10.1016/j.msea.2017.02.028 Received 15 December 2016; Received in revised form 6 February 2017; Accepted 7 February 2017 Available online 08 February 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
the cans filled with TiAl powders were compacted via HIP technique at 1533 K and 160 MPa for 4 h with heating rate of about 4 K per min, and then furnace cooled to room temperature. The stainless steel containers were removed by lathe machining and consolidated bulk samples were obtained (designated as HIPed sample). In order to generate duplex or lamellar structure, the HIPed samples were annealed at 1553 K for 4 h or 1633 K for 0.5 h in vacuum, with furnace cooling in both conditions (designated as Dsep and Lsep samples, respectively). For IHH processing route, the first HIP stage had the same parameters as that in SHH route. The difference is that at the end of HIP process, temperature were straight elevated to 1553 or 1633 K and maintained for 4 or 0.5 h to generate duplex or lamellar structure, followed by furnace cooling in both conditions (named as Dint and Lint samples, respectively). The gauge size of tested samples is ϕ8 mm × 40 mm , and the tensile tests were carried out using MTS Landmark testing system with uniaxial quasistatic strain rate of 2 × 10−4 s−1 at room temperature. The microstructure was investigated by backscattered electronic (BSE) imaging using a field emission gun scanning electron microscope. Specimens for BSE investigation were electro-polished using a solution of 95% ethylalcohol and 5% perchloric acid (HClO4) at 253 K with voltage of 40 V.
stresses, which make the sample or component defective prior to testing or service. Accordingly, the modifications in traditional heat treatment processes to suppress the formation of defects and produce the defect-free samples or components would be significant step towards enhancing mechanical properties in TiAl alloys. Atmospheric pressure during heat treatment has been developed in special field to recover the mechanical properties of turbine blades in industry [23,24]. With a hot isostatic pressing atmosphere, the microvoids or microcracks could be reclosed and thus the rejuvenated mechanical properties would be obtained. Inspired by the advantages of pressure to heat treatment, we have taken the hot isostatic pressing (HIP) technique to provide a pressure environment during heat treatment to suppress the potential formation of microcracks in TiAl intermetallic alloys. Based on HIP technique, a special processing method integrating HIP and heat treatment (IHH) was designed. For comparison, traditional separated HIP and subsequent heat treatment (SHH) in vacuum was also conducted. The present study explores the economical strategy for improving the mechanical properties of TiAl alloys, especially for components with relatively complex geometries and large-scale sizes. The mechanical properties for samples generated by two different methods were tested and analyzed, and corresponding microstructural evolutions as well as defect-induced deterioration in mechanical properties were investigated in detail.
3. Results 2. Experimental procedures 3.1. Mechanical properties The TiAl alloy used in this study has a nominal chemical composition (at%) of 47Al, 2Cr, 2Nb and Ti in balance. Alloy ingots were first prepared by arc melting and drop casting into Cu molds using commercially pure metals. The ingots were then remelted and atomized in argon atmosphere to generate alloy powders with mean particle size of about 100 µm. Powders were filled into 304 stainless steel cans with diameter of 60 mm and 150 mm in length, followed by degassing at room temperature, 623 K and 873 K for 1, 2, and 4 h, respectively. After degassing, cans were sealed in vacuum of about 10−1 Pa, which will be used for HIP and heat treatment. Two different processing routes were then conducted respectively, which are schematically illustrated in Fig. 1. During SHH processes,
Fig. 2 exhibits the tensile properties of TiAl samples produced by different processing routes. The typical engineering stress‒strain curves are shown in Fig. 2(a) and (b). The mean yield stress and elongation to failure of HIPed samples are measured as 480 ± 21 MPa and 2 ± 0.3%. When subjected to annealing at 1553 K to form duplex microstructure via SHH route, the Dsep samples have an average yield stress and elongation to failure of 420 ± 18 MPa and 2.4 ± 0.4%, while Dint samples produced by IHH route have higher yield stress and elongation to failure up to 540 ± 25 MPa and 3 ± 0.3%. When it comes to Lsep samples with lamellar structure which were annealed at 1633 K, as exhibited in Fig. 2(b), the yield stress and elongation to failure are measured as 550 ± 28 MPa and 0.6 ± 0.1%, respectively. In contrast, the Lint samples treated by IHH route have higher yield stress and elongation to failure up to 650 ± 30 MPa and 2.1 ± 0.2%. According to the results of mechanical properties, the HIPed samples have moderate yield strength and tensile ductility. When the traditional heat treatment in vacuum was applied to generate the lamellar or duplex microstructure, it is hard to achieve a good combination between the strength and tensile ductility. Taking the lamellar structure as an example, the yield stress could be improved; however, such improvement is accompanied by a significant sacrifice of tensile ductility. In contrast, when processed through IHH route, the yield stress and elongation to failure of both duplex and lamellar structures are increased simultaneously. Corresponding to Fig. 2(a) and (b), the true stress and strain hardening rate were plotted against true strain in Fig. 2(c) and (d). As compared with the Dsep sample, the Dint sample produced by IHH route shows higher strain hardening rate during tensile deformation, indicating stronger strain hardening ability and hence enhanced tensile ductility, which is in line with the experimental results exhibited in Fig. 2(a). For the lamellar structure, the Lsep sample is quickly failed at early stage of tensile deformation, while Lint sample sustains to relatively higher tensile strain. According to Considère theory, plastic instability during uniaxial tensile deformation sets in as strain hardening rate decreases to the same level as true stress [25]. However, all samples do not follow the prediction of Considère criterion. Instead, failures all occurred at low strains where the strain hardening rates are still much greater than true stresses, which means that the failure of
Fig. 1. Schematic illustration of two different processing routes. (a) Traditionally separated route (SHH) HIPing first at 1533 K and 160 MPa for 4 h, followed by heat treatment in vacuum at 1633 and 1553 K for 0.5 and 4 h, respectively. (b) Integrated processing route (IHH) undertaking heat treatment immediately at the end of HIP with pressure atmosphere of 160 MPa.
372
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 2. Mechanical properties of TiAl alloy. (a) Engineering stress‒strain curves for duplex microstructure generated by IHH (Dint) and SHH (Dsep). The curve of HIPed sample was also plotted for comparison. (b) Engineering stress‒strain curves for lamellar microstructure generated by IHH (Lint) and SHH (Lsep) routes. Corresponding true stress (σt) and strain hardening rate (Θ) plotted against the true strain for (c) duplex and (d) lamellar structure, respectively.
to the gray contrast of micrograph, some Al atoms have segregated in γ grains, as denoted by the yellow asterisks [27], while some heavy atoms (e.g. Nb or Cr) have segregated in lamellar regions, which are denoted by the red asterisks. The strong atom segregation implies that the HIPed sample is far from equilibrium state, which is one of the reasons to accelerate the solid state phase transformation during heat treatment stage. Fig. 4 presents the micrographs of duplex structure generated by two different processing routes. Fig. 4(a) shows the duplex structure of Dsep sample, which consists of nearly equiaxed γ grains and lamellar α2/γ domains in about half and half proportion. Fig. 4(b) is the magnified micrograph corresponding to encircled area in Fig. 4(a). Some microcracks are indicated by white arrows in Fig. 4(a) and (b), with length ranging from several 10–100 s of microns. In contrast to Dsep samples, Dint samples produced by IHH route exhibit duplex structure free of microcrack. Fig. 4(c) shows a typical microstructure of Dint sample, and the area marked by rectangular line is enlarged and shown in Fig. 4(d). The area fractions of lamellar colonies were counted as about 41% and 43% in Dsep and Dint samples. In order to verify that the Dint sample is free of microcracks, more than 70% cross-sectional area had been screened, and no visible cracks were detected during screening. Based on BSE micrographs, the sizes of different microstructural characteristics were counted and their statistical distributions are shown in Fig. 5. For Dsep sample produced by SHH route, Fig. 5(a)– (c) display the size distributions of γ grain, α2/γ domain and α2/γ lamellar spacing, and their average sizes are counted as 55.7, 61.9 and 2.5 µm, respectively. For comparison, corresponding size distributions of Dint sample produced by IHH route are displayed in Fig. 5(d)–(f), and their mean size are 43.7, 69.6 and 2.3 µm. The inset in Fig. 5(c) illustrates schematically the definition of lamellar spacing, which equals to the sum of mean thickness of α2 and γ lamellae. According to statistical results of characteristic size in lamellar structure, one can conclude that the isostatic pressure during heat treatment stage of IHH processing has little influence on the overall
TiAl alloy is attributed to the premature fracture, rather than its intrinsic poor mechanical properties [26]. Hence, approaches to suppress the premature failure would be beneficial to improve the mechanical properties of TiAl alloy.
3.2. Microstructure observation BSE micrographs revealed the influence of processing routes on the evolution of microstructures for TiAl alloy. Fig. 3 shows the micrograph of HIPed sample, which is extremely heterogeneous in microstructure. The regions of equiaxed γ grains account for about 45% in area fraction, and the rest are mainly fine stripe-like colonies. According
Fig. 3. BSE micrograph of HIPed sample consisting of equiaxed γ-TiAl grains and stripelike colonies. As denoted by asterisks (yellow and red ones correspond to Al- and Ti-rich regions, respectively), atom segregations are shown in both equiaxed and stripe-like regions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
373
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 4. BSE micrographs of duplex structure generated by SHH and IHH routes, consisting of lamellar colonies and equiaxed γ grains. (a) Micrograph correlated to SHH route showing microcracks as marked by arrows. (b) Corresponding high magnification image of the enclosed area in (a). (c) Micrograph without cracks for samples produced by IHH route. (d) Enlarged view of corresponding enclosed area in 4(c) indicating defect-free duplex structure. (α2 lamellae in lamellar colony is distinguished by brighter contrast due to its higher concentration of Ti).
Fig. 5. Statistical distribution of different microstructural sizes for duplex structure. Size distributions of (a) γ grain, (b) α2/γ domain and (c) lamellar spacing for Dsep sample produced by SHH route. Corresponding size distributions of (d) γ grain, (e) α2/γ domain and (f) lamellar spacing for Dint sample produced by IHH route.
374
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
Fig. 6. BSE micrographs of lamellar structure generated by SHH and IHH routes. (a) Micrograph correlated to SHH route showing microcracks (marked by white arrows). (b) High magnification image corresponding to enclosed area of (a) exhibiting defective lamellar structure. (c) Micrograph free of cracks for samples generated by IHH route. (d) Enlarged view of corresponding enclosed area in (c).
Fig. 7. Statistical distribution of different microstructural sizes for lamellar TiAl. Size distributions of (a) α2/γ domain and (b) lamellar spacing for lamellar structure correlated to SHH route. Corresponding size distributions of (c) α2/γ domain and (d) lamellar spacing for lamellar structure generated by IHH route.
375
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
treatment without exterior pressure would break the equilibrium condition and thus possibly result in swelling in microstructure [30]. For TiAl alloys produced by powder metallurgical methods, the subsequent heat treatment is always necessary and important to adjust the microstructure for optimized lamellar or duplex morphology. Therefore, the swelling tendency during heat treatment stages without atmospheric pressure will increase the possibility of debonding and cracking. Additionally, the heat treatment process involves substantial solid phase transformation for TiAl alloys, during which the displacive transformation [31] could result in both microscopic and macroscopic stresses [32]. Accordingly, during the process of heat treatment, under the internal stress resulting from the breaking of equilibrium in pores and the solid phase transformation, the weak-bonding area as a result of load shielding would be failed and then the cracking will occur. Based on the microstructure observations, both lamellar spacing and phase fractions are not obviously changed due to isostatic pressure at heat treatment stages in either duplex (Fig. 5) or lamellar (Fig. 7) microstructures. Therefore, in present investigation, the isostatic pressure during heat treatment will mainly sustain the equilibrium in pores, and provide an exterior force to balance the potential internal stresses due to phase transformation. As a result, the potential cracking was suppressed and the crack-free microstructure was obtained finally in TiAl alloy.
evolution of microstructure. As shown in Fig. 5, the Dint sample has similar microstructural sizes in terms of γ grain, lamellar colony and lamellar spacing, as those in Dsep sample. However, it should be noted the area fraction of lamellar colonies is slightly improved in Dint sample, and the lamellar spacing size in Dint sample is a little finer than that in Dsep sample. It seems that the pressure during IHH processing has a slight positive effect on the formation of lamellar structure. Fig. 6 shows the micrographs of lamellar structure generated by different processing routes. As shown by Fig. 6(a), some microcracks are indicated by white arrows, and the enclosed area is further displayed as enlarged view in Fig. 6(b). The microcracks indicate that Lsep samples which were produced by SHH route are defective in microstructure prior to tensile deformation. However, as shown by Fig. 6(c) and (d), the Lint sample produced by IHH route shows lamellar structure free of microcracks, indicating successful suppression of crack formation due to isostatic pressure. Additionally, more than 70% cross-sectional area of Lint sample had been screened and no visible cracks were detected. Fig. 7 shows statistical distributions of lamellar domain size and lamellar spacing for Lsep and Lint samples. Fig. 7(a) and (b) display the distribution of domain size and lamellar spacing of Lsep sample, while corresponding distributions of Lint sample are depicted in Fig. 7(c) and (d). The mean lamellar domain sizes in Lsep and Lint sample are close to each other, which equal to 75.4 and 70.4 µm, respectively. Similar to the size of lamellar domain, the average lamellar spacing between Lsep and Lint sample has no obvious difference in magnitude. According to Figs. 6 and 7, it could be concluded that isostatic pressure during IHH route has little influence on the overall evolution of microstructure, but strongly suppressed the formation of microcracks in lamellar structure. Therefore, the optimized mechanical properties of samples produced by IHH route vis-à-vis those processed by SHH route are mainly attributed to their crack-free microstructures.
4.2. Influence of microcrack on mechanical properties The mechanical properties of intermetallic alloys are extremely sensitive to defects in microstructure. With regards to TiAl alloys which have relatively high brittle-to-ductile transition temperatures [33,34], microcracks are extremely harmful in deteriorating mechanical performance at room or even at low temperatures. Generally, microcrack will decrease the tensile elongations of materials with low ductility, since their strain hardening will be suppressed and the samples are mostly elongated via crack propagation rather than traditional plastic straining. That is the reason why many intermetallics like TiAl alloy always exhibit negligible tensile elongation. However, the influence of initial defects such as pore or microcrack to the yield strength seems not quite as much as that to the tensile elongation, which is anomalous to the current results where yield strength is also decreased due to microcracks (Fig. 2(a) and (b)). For example, the commercially pure Ti with different porosities [35] and the Al-Si alloy with different intensity of damage [36] have shown significant reduction of tensile elongation but little change of yield strength with increase of defects. For current TiAl alloy, however, both yield strength and tensile elongation will be decreased if microcracks exist within the microstructure. The primary reason for the difference between present study and literatures [35,36] might be that such materials are more ductile, in which the propagation of defects will be initiated after the dislocationdominated yielding stage. However, during tensile deformation at room temperature for TiAl alloy, the nucleation of cracks had been detected even prior to macroscopic yielding by acoustic emission technique [22,37], manifesting that the cracking behavior had happened along with the dislocation gliding during yielding process. The microcracks can be stable at least within the size comparable to the colony size or grain diameter [10], and thus the tensile process will involve the dislocation and cracking behavior at the same time. Therefore, the yield strength is not only determined by the activation of dislocation gliding, but also the morphology and geometry of microcracks. The microcrack will directly reduce the effective crosssectional area to bear the applied load, leading to decrease of yield strength. In addition, similar to the deformation in composites, the propagation of crack could produce a strain besides the traditional elastic and plastic strains, lowering down the flow stress curves and showing lower apparent yield stresses. Since the brittle materials are more sensitive to crack behavior in mechanical property, both their yield strength and tensile elongation will be deteriorated due to
4. Discussion 4.1. Benefits of isostatic pressure to crack-free microstructure During hot isostatic pressing, the metallic powders inside the can will deform, diffuse and finally be consolidated into the desired solid geometry with the thermal and dynamic assistance of temperature and pressure. However, it is still difficult in reality to remove the artifacts from final bulk samples or components, which is therefore harmful to their mechanical performance, especially for materials with low ductility. Load shielding is one of the possible reasons resulting in defects in microstructure. As proposed by Wadley et al. in their investigation, can shielding will decrease the pressure supported by powder, making the real pressure exerted on powder less than that exerted on can, especially at relatively low temperature or when the can has a yield strength comparable to or greater than that of the powder [28]. In addition, for cylindrical can with greater size in length than that in diameter, the load shielding is more severe along the axis direction than that along the radius direction [29], indicating the size dependence of load shielding in magnitude. Therefore, the bigger the sample or component is, the stronger the load shielding will become. For present investigation, the 304 stainless steel was chose as the can material, which will lower down the can shielding effect due to its low flow stress at HIP temperature. However, the relatively large size in diameter and especially in length would aggravate the load shielding, which decreases the real pressure exerted on powder and thus increase the probability to produce pores and weak bonding. Furthermore, since the can interior is not in absolute vacuum condition, the pores at final stage of HIP will be gas-filled due to the concentration of gas molecules, and the equilibrium pressure inside will then make such pores difficult to be closed. Even if the pores would be closed finally by HIP and give a full density, the subsequent heat 376
Materials Science & Engineering A 688 (2017) 371–377
L. Chen et al.
References
microcrack. The present result is in alignment with the investigation in low-ductile Ti-6Al-4V alloys with different porosities [35], where both yield strength and tensile elongation were decreased with the increase of defects. Accordingly, the microcracks should be the primary reason for the low yield strength and tensile ductility for duplex and lamellar samples produced by SHH route. Therefore, special integrated processing route with isostatic pressure during entire process has benefits to suppress the formation of microcrack in both duplex and lamellar microstructures, and provides improved combination between yield strength and tensile elongation.
[1] F. Appel, H. Clemens, F.D. Fischer, Prog. Mater. Sci. 81 (2016) 55–124. [2] F. Appel, J.D.H. Paul, M. Oehring, Gamma Titanium Aluminide Alloys: Science and Technology, John Wiley & Sons, Weinheim, 2011. [3] Y.W. Kim, JOM 47 (1995) 39–42. [4] Y.W. Kim, D.M. Dimiduk, JOM 43 (1991) 40–47. [5] M. Yamaguchi, H. Inui, K. Ito, Acta Mater. 48 (2000) 307–322. [6] W. Wallgram, T. Schmölzer, L. Cha, G. Das, V. Güther, H. Clemens, Int. J. Mater. Res. 100 (2009) 1021–1030. [7] C.T. Liu, P.J. Maziasz, J.L. Wright, MRS Proceedings, Cambridge Univ Press, 1996, pp. 83. [8] C.T. Liu, J.H. Schneibel, P.J. Maziasz, J.L. Wright, D.S. Easton, Intermetallics 4 (1996) 429–440. [9] Y. Wu, S.K. Hwang, Acta Mater. 50 (2002) 1479–1493. [10] C.T. Liu, P.J. Maziasz, Intermetallics 6 (1998) 653–661. [11] Y. Song, D.S. Xu, R. Yang, D. Li, Z.Q. Hu, Intermetallics 6 (1998) 157–165. [12] F. Appel, M. Oehring, R. Wagner, Intermetallics 8 (2000) 1283–1312. [13] P. Au, Carleton University, Ottawa, 1997. [14] T. Tetsui, K. Shindo, S. Kobayashi, M. Takeyama, Scr. Mater. 47 (2002) 399–403. [15] Z.C. Liu, J.P. Lin, S.J. Li, G.L. Chen, Intermetallics 10 (2002) 653–659. [16] J.N. Wang, J. Yang, Q. Xia, Y. Wang, Mater. Sci. Eng. A 329–331 (2002) 118–123. [17] E. Schwaighofer, H. Clemens, S. Mayer, J. Lindemann, J. Klose, W. Smarsly, V. Güther, Intermetallics 44 (2014) 128–140. [18] H. Clemens, W. Wallgram, S. Kremmer, V. Güther, A. Otto, A. Bartels, Adv. Eng. Mater. 10 (2008) 707–713. [19] A. Couret, G. Molénat, J. Galy, M. Thomas, Intermetallics 16 (2008) 1134–1141. [20] H.A. Calderon, V. Garibay-Febles, M. Umemoto, M. Yamaguchi, Mater. Sci. Eng. A 329–331 (2002) 196–205. [21] K. Edalati, S. Toh, H. Iwaoka, M. Watanabe, Z. Horita, D. Kashioka, K. Kishida, H. Inui, Scr. Mater. 67 (2012) 814–817. [22] D. Hu, A. Huang, H. Jiang, N. Mota-Solis, X. Wu, Intermetallics 14 (2006) 82–90. [23] A. Baldan, J. Mater. Sci. 26 (1991) 3409–3421. [24] A.G. Peterson, D.W. Gandy, G. Frederick, J.T. Stover, R. Viswanathan, Google Patents, 2002. [25] X.L. Wu, M.X. Yang, F.P. Yuan, G.L. Wu, Y.J. Wei, X.X. Huang, Y.T. Zhu, Proc. Natl. Acad. Sci. USA 112 (2015) 14501–14505. [26] Y.G. Kim, J.M. Han, J.S. Lee, Mater. Sci. Eng. A 114 (1989) 51–59. [27] E. Schwaighofer, B. Rashkova, H. Clemens, A. Stark, S. Mayer, Intermetallics 46 (2014) 173–184. [28] H.N.G. Wadley, R.J. Schaefer, A.H. Kahn, M.F. Ashby, R.B. Clough, Y. Geffen, J.J. Wlassich, Acta Mater. 39 (1991) 979–986. [29] J. Xu, R.M. McMeeking, Int. J. Mech. Sci. 34 (1992) 167–174. [30] H.V. Atkinson, S. Davies, Metall. Mater. Trans. A 31 (2000) 2981–3000. [31] Y. Sun, Philos. Mag. Lett. 78 (1998) 297–305. [32] B. Verlinden, J. Driver, I. Samajdar, R.D. Doherty, Thermo-Mechanical Processing of Metallic Materials, Elsevier, Amsterdam, 2007. [33] C.T. Liu, Scr. Metall. Mater. 27 (1992) 599–603. [34] D.W. Hu, Rare Met. 35 (2016) 1–14. [35] R. Bourcier, D. Koss, R. Smelser, O. Richmond, Acta Mater. 34 (1986) 2443–2453. [36] G. Guiglionda, W. Poole, Mater. Sci. Eng. A 336 (2002) 159–169. [37] R. Botten, X. Wu, D. Hu, M.H. Loretto, Acta Mater. 49 (2001) 1687–1691.
5. Conclusions A special method integrating traditional hot isostatic pressing and heat treatment was developed to improve the mechanical performance of TiAl alloy. The main conclusions are drawn as follows: (1) An integrated processing method has been developed which combines the traditional HIP and heat treatment process. With this method, an isostatic atmospheric pressure is introduced during entire process from consolidation of powder to heat treatment of coupon. (2) According to tensile testing results, both samples with lamellar and duplex microstructure generated by integrated processing route have simultaneous improvement in yield stress and elongation to failure. Compared with samples processed by traditionally separated HIP and heat treatment, the yield stress and elongation to failure of lamellar structure have been improved from 550 ± 28 MPa and 0.6 ± 0.1%, to 650 ± 30 MPa and 2.1 ± 0.2%. While in the duplex structure, they are increased from 420 ± 18 MPa and 2.4 ± 0.4%, to 540 ± 25 MPa and 3 ± 0.3%, respectively. (3) As evidenced by microstructural investigation, the microcracks play an important role in deteriorating mechanical properties of TiAl alloy. By using integrated processing route, both lamellar and duplex structures free of microcrack are generated, and thus an improvement in mechanical properties has been achieved. Acknowledgements The authors would like to thank the financial support by Natural Science Foundation of China with Grant no. 51301187.
377