Effect of isothermal forging strain rate on microstructures and mechanical properties of BT25y titanium alloy

Effect of isothermal forging strain rate on microstructures and mechanical properties of BT25y titanium alloy

Materials Science & Engineering A 673 (2016) 355–361 Contents lists available at ScienceDirect Materials Science & Engineering A journal homepage: w...

3MB Sizes 2 Downloads 62 Views

Materials Science & Engineering A 673 (2016) 355–361

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Effect of isothermal forging strain rate on microstructures and mechanical properties of BT25y titanium alloy Xuemei Yang a,n, Hongzhen Guo a, Zekun Yao a, Shichong Yuan b a b

School of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an 710072, PR China China National Erzhong Group Co., Deyang 618013, PR China

art ic l e i nf o

a b s t r a c t

Article history: Received 31 May 2016 Received in revised form 19 July 2016 Accepted 19 July 2016 Available online 20 July 2016

In order to confirm the optimal isothermal forging process for BT25y alloy, the effect of strain rate on the microstructure and mechanical properties was studied. Quantitative relationships between microstructure characteristics and mechanical properties were obtained. Results showed that as the strain rate decreased, the lamellar α coarsened, the β grain boundary connected by small recrystallized grains disappeared, and the microstructure became much more uniform. Tensile strength for the three processes all maintained at a high level, but the elongation tended to first increase and then decrease with dropping strain rate. The tensile fracture changed from quasi-cleavage fracture to complete ductile fracture, and finally transformed into a mixture of ductile fracture and brittle fracture at lower strain rate. Fracture toughness improved with increased strain rate, but the thermal stability deteriorated gradually. & 2016 Elsevier B.V. All rights reserved.

Keywords: BT25y alloy Isothermal forging Strain rate Microstructure Mechanical property

1. Introduction Titanium alloys are getting widely applications in aerospace industry for their excellent mechanical properties like low density, high specific strength, good corrosion resistance and elevated service temperature [1,2]. However, with the rapid development of aero engines, higher requests for elevated-temperature strength and thermal-mechanical stability have been raised for working under extreme environment [3]. Thus an increasing number of studies on high temperature titanium alloys have been performed, and BT25y alloy is one of the satisfying alloys. BT25y alloy is developed by adjusting the content of elements like Mo, Zr in BT25 alloy so as to improve its heat strength and high temperature properties. Its nominal composition is Ti-6.5Al2Sn-4Zr-4Mo-1 W-0.2Si [4]. The alloy is recommended as the material for making high pressure compressor and rotor blade working at 450–550 °C [5]. But domestic study of BT25y titanium alloy is still in the stage of engineering test. Generally speaking, the microstructure of titanium alloys is very sensitive to strain rate during hot deformation. Quantities of researches have been conducted to study the effect of strain rate on microstructure or mechanical properties of titanium alloys in recent years. Z.X. Du [6] investigated the strain rate dependence of microstructural evolution in β titanium alloy during subtransus n

Corresponding author. E-mail address: [email protected] (X. Yang).

http://dx.doi.org/10.1016/j.msea.2016.07.084 0921-5093/& 2016 Elsevier B.V. All rights reserved.

superplastic deformation, and found that higher strain rate deformation resulted in elongated α and β phases, while deformation at lower strain rate led to growth of α grain and spheroidization of β phase. T. B. Wang [7] explored the effect of strain rate on deformation twinning behavior of α-titanium, and con¯ , cluded that the high-speed deformation twins included 1122 ¯ ¯ , 1121 ¯ contraction twins and 1012 extension twins in 1124

{

{

}

{

}{

}

}

α-titanium, but the total content of twin increased sharply after

high-speed compression. D. He [8] discussed the influence of deformation strain rate on the Burgers orientation relationship and variants morphology during β-α phase transformation in a near α titanium alloy, and observed that the length of α lamellae decreased with increasing strain rate, but strain rate had no significant influence on the obeying of Burgers orientation rule. J. Zhang [9] investigated the effect of strain rate on the tension behavior of Ti-6.6Al-3.3Mo-1.8Zr-0.29Si alloy at low temperatures, and noticed that the value of initial yield stress increased with increasing strain rate, but the isothermal strain hardening behavior changed little with different strain rates. C. Li [10] considered the effect of strain rate on stress-induced martensitic formation and the compressive properties of Ti-V-(Cr, Fe)-Al alloys, and pointed out that the triggering stress for stress-induced martensitic transformation and failure strain increased continuously with increasing strain rate, but the compression strength was found to first increase and then decrease with increasing strain rate. S.Q. Wang [11] studied the effect of strain rate on the tensile properties, strain hardening behavior, strain rate sensitivity, and fracture

356

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

characteristics of a dissimilar joint between Ti-6Al-4V and Ti17 alloys, and found that both hardening capacity and strain hardening exponent decreased with increasing strain rate. The deformation strain rate is a critical factor to control microstructures and corresponding mechanical properties of titanium alloys during hot deformation. However, the effect of strain rate on microstructures and mechanical properties of BT25y titanium alloy during hot deformation has not been performed yet. The purpose of this experiment is to research on the effect of deformation strain rate on the microstructures and mechanical properties of BT25y alloy, and establish the quantitative relationship between microstructure parameters and mechanical properties for precisely control of forging processes.

Fig. 1. The initial microstructure of BT25y alloy bar.

2. Experimental The raw material is provided in the form of alloy bar with a diameter of 270 mm, whose chemical components are listed in Table 1. The β transus temperature is determined to be 965 °C by metallographic technique. The original microstructure is presented in Fig. 1. As shown in the figure, the microstructure is mainly composed of equiaxed α grain and β matrix, and a small amount of lamellar α can be observed in local region. It consists of 50% primary α phase with the average grain size of 7 mm and transformed β with the secondary lamellar α thickness of 1.1 mm. The isothermal forging is performed on a four-column hydraulic press with the maximum working load of 6300 kN, and strain rate is precisely controlled by the electric control system. The forging mold and temperature control device are self-made. Since the sustainable maximum temperature that the heating furnace can provided is 930 °C, the die temperature is determined to be 930 °C. Billet preheating and heat treatment are performed on the SX-10–13 chamber electric furnace. The billet preheating rate is 0.8 min/mm [12]. Previous tests have confirmed the optimal deformation temperature and heat treatment regime are 960 °C and 900 °C/1 h, AC þ600 °C/6 h, AC, respectively. Thus, the experiment scheme for strain rate optimization is shown in Table 2. Blank size for fracture toughness tests is 43 mm (length)  41 mm (width)  56 mm (high), and blank size used for other performance tests is 60 mm (length)  21 mm (width)  30 mm (high). In addition, the height direction (namely, the isothermal compression direction) should be parallel to the center axis of original bar. Room temperature (RT) and high temperature (HT) tensile tests are performed on the ENST-1196 tensile machine. Fracture toughness (KIC) tests are performed by the compact tensile method on INSTRON-1251 test machine. Microstructure observation is conducted on the OLYMPUSPM-G3 optical microscope (OM). Optical quantitative measurement is completed by the Image-Pro Plus (IPP) metallographic image analysis software, and the average value measured from 8 view fields is taken as the characteristic microstructure parameter. Subtle morphology and dislocation distribution are observed on the Tecnai F30 G2 field emission TEM. Fracture surfaces of tensile specimens are observed using MIRA3 TESCAN field emission SEM. Table 1 Chemical composition of BT25y alloy bar. Element Al Wt/%

Mo

Zr

Sn

W

Si

O

C

N

H

Ti

6.44 3.94 3.62 1.68 0.58 0.22 0.09 0.046 0.01 0.001 Bal

Table 2 Processing programs of isothermal forging for BT25y alloy. Pro. Strain rate/s  1 Deformation degree/% 1

1  10  1

2 3

1  10  2 1  10  3

60

Billet temperature/°C

Die tempera- Heat ture/°C treatment

960

930

900 °C/ 1 h, AC þ 600 °C/ 6 h, AC

3. Results and discussion 3.1. Microstructural evolution Fig. 2 shows the microstructure after different isothermal forging processes and identical heat treatment condition. It can be observed that the strain rate of isothermal compression has a significant influence on the microstructure of BT25y alloy. Since the deformation temperature approaches to the phase transition point of BT25y alloy, the microstructure corresponding to three processes almost all belongs to basket-weave structure. For process 1 (Pro.1), primary equiaxial α almost disappears due to the deformation heat at high strain rate. Large β grains form in the microstructure, and secondary acicular α distributes interlacing with each other. As can be seen from Fig. 2(a), the β grain boundaries are greatly broken and composed of small recrystallized grains. When deformed at high strain rate, large amounts of dislocation multiplication (shown in Fig. 3) will provide more nucleation sites for recrystallization. Moreover, impurities and defects at the grain boundary give rise to more stored energy so that a great number of recrystallized grains generate [13]. For process 2 (Pro.2), thick intermittent or continuous grain boundary with a certain degree of distortion can be observed. As can be seen from Fig. 2(b), large numbers of equiaxial grains distributed in the same orientation appear near the grain boundary. A large number of secondary lamellar α precipitates at the grain boundaries and spreads to the intracrystalline during the transport process before isothermal forging. Then spheroidization of α lamellae occurs for the reduction of surface energy during deformation [14]. Actually, large deformation induces inner-stress such as shear stress and dislocation pile-up, which can provide enough energy to start dislocation gliding and atomic diffusion. Then the β phase penetrates straightly into α phase, and partially or fully separates α lamellae into shorter segments, thus small equiaxial grain forms. Fig. 4 shows the sub-grain boundaries formed across the lamellar α.

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

357

Fig. 3. TEM image of the specimen deformed at the strain rate of 1  10  1 s  1.

Fig. 4. Sub-grain boundary formed within the α lamellae.

Fig. 2. The microstructure of BT25y alloy under deformation strain rates of (a) 1  10  1 s  1; (b) 1  10  2 s  1; (c) 1  10  3 s  1.

boundary sliding and diffusion creep, which causes growth of recrystallization grains and coarsening of lamellar alpha. 3.2. Quantitative analysis of relationship between mechanical properties and microstructures

For process 3 (Pro.3), with extended deformation time at low strain rate, the intracrystalline lamellar α becomes much thicker. It is difficult to distinguish grain boundary and grain interior so as to obtain a uniform microstructure. Recrystallized equiaxial grains can almost not be observed. There are two reasons for unobvious recrystallization under low strain rate. One is less nucleation of recrystallization grains. Deformation at small strain rate may lead to accelerated decline of dislocation density, and timely release of deformation heat, which will result in less crystal defects and stored energy, consequently the tendency for recrystallization reduces. The other is the growth-up and further deformation of recrystallized grains. The corresponding deformation time for process 1, 2, 3 are 15 min 16 s, 91 s, 9 s, respectively. Long period of high temperature deformation under low strain rate helps grain

Tensile property values of BT25y alloy after hot deformation are listed in Table 3, and the data are average of two experimental samples. Fig. 5 has shown the variation trend of RT tensile strength, plasticity and microstructure parameters with strain rate. The forging strain rate has a remarkable influence on the tensile properties of BT25y alloy. As can be seen from Table 3, tensile strength and plasticity are better matched under lower strain rate. However, when deformed at high strain rate, tensile strength slightly increases, but tensile plasticity drops a lot. For Pro.1, due to the increase of phase boundary area, it needs larger applied stress when dislocations move from β matrix to α platelets for further deformation. As can be seen from Fig. 6, a lot of dislocations pile up at the phase boundary. Besides, impurities

358

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

Table 3 Tensile properties of BT25y alloy at different deformation strain rates. Pro.

Test temp. /°C

UTS sb/MPa

YS ss/MPa

Elongation δ/%

Reduction to failure Ψ/%

1

RT 550 RT 550 RT 550

1193.5 870 1164 857.5 1187.5 840

1098 712.5 1069 710 1114.5 712.5

10 13 14.2 18 14 17

17 26 19 31.5 22.5 42

2 3

and defects enrichment at grain boundaries impede the dislocation motion so that the tensile strength of Pro.1 is a little higher than that for other processes. There's almost no distinctive grain boundary and recrystallization grain can be observed from the microstructure of Pro.3, and the relatively uniform microstructure is good for the improvement of tensile properties. It can be obtained from Fig. 5 that when deformed at small strain rate, both volume fraction and width of lamellar α maintain at a quite high level. The α phase with hexagonal close-packed (HCP) structure is generally thought to be tough phase, which can provide considerably resistance to dislocation movement, thus tensile strength increases. For Pro.1, the un-uniform microstructure would abate the deformation coordination ability. Uneven plastic deformation and local stress concentration happen easily, so that cracks appear much earlier. In addition, the cross-distribution of slender α on the β matrix can bring about the pile-up of interfacial dislocation, which can easily cut off lamellar α (shown in Fig. 7) and promote further extension of micro cracks, thus the tensile plasticity reduces. As can be seen from Fig. 2, due to the prolonged deformation time under small strain rate, the original slender α grows into thick strip α. The thick strip α can not be easily cut off, which forces the alteration of dislocation glide direction, therefore stress concentration reduces and fracture time extends, hence specimens in Pro.3 present high plasticity. Besides, some scholars [15,16] point out that tensile plasticity is related to the thickness of α layer in β grain boundary. Once grain boundary α becomes thinner, pile-up amounts of dislocation at grain boundary will be less, and tensile plasticity is better. From the HT tensile properties listed in Table 3, it can be found that HT strength drops a lot compared with that in RT. Nevertheless, strength difference for various strain rates is not obvious, where ultimate strength fluctuates narrowly around 855 MPa, and yield strength maintains at approximately 710 MPa. The variation trend of HT plasticity with strain rate is the same as the RT plasticity. However, the HT plasticity is higher than that in RT since the deformation mechanisms corresponding to these two conditions

Fig. 6. Dislocation pile-up at the phase boundary in specimen deformed at strain rate of 1  10  1 s  1.

Fig. 7. Dislocation gliding across lamellar α.

Fig. 5. Variation of RT tensile properties and microstructure parameters with strain rate: (a) strength; (b) plasticity.

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

Fig. 8. Variation of fracture toughness and microstructure parameters with strain rate.

are different [17]. As tensile temperature rises, surface free energy of atoms and thermal activation energy of materials promote so that critical shear stress for plastic slipping reduces, thus the HT tensile strength drops a lot [18]. With the increase of tensile temperature, grain boundary/phase boundary sliding becomes easier [19,20]. Thus the coordinate deformation through grain boundary/phase boundary sliding turns out to be the dominant deformation mechanism at HT, which relaxes the mainly affecting factor of crack nucleation – stress concentration, and made deformation go on more smoothly. Actually, crack nucleation resistance and crack propagation resistance are two major factors influencing plasticity [21]. Accordingly, the HT tensile plasticity improves a lot compared with the RT plasticity. Fig. 8 shows the variation trend of fracture toughness and microstructure parameters with strain rate of BT25y alloy. It can be seen from the figure that the fracture toughness of Pro.2 is a little higher than that of Pro.1, but fracture toughness of Pro.3 is greatly reduced. As can be seen from Fig. 2, lamellar α coarsens obviously with the decrease of strain rate, which indicates that thickness of lamellar α is an important factor for the fracture toughness of BT25y alloy. Fig. 9(a) shows that the crack tends to propagate along thick α lamellae, but it will cut off the thin α platelet to move forward. Thus, thick lamellar α can not be easily cut off and will increase the tortuousness of crack propagation path. Generally speaking, fracture toughness improves with the thickening of α platelets [22], but thickness of lamellar α exceeding a certain threshold will also result in a loss of fracture toughness. The low fracture toughness of Pro.3 with thick α platelets is consistent with above conclusion. Crack is prone to extend along the path with

359

least resistance [23]. Densely interlaced lamellar α colonies make the crack propagation alter its direction continually, which leads to winding crack path with lots of branches, thus fracture toughness improves [24]. However, when lamellar α with excessive thickness participates in the deformation, resistance for dislocation motion in single lamellar α is weakened so that dislocation can move forward easily within α platelet. Moreover, when dislocation moves from β matrix to thick lamellar α, it will pile up at the phase boundaries. Such phase boundaries of thick lamellar α would turn into a source of micro cracks and reduce the fracture toughness [25]. Thus, fracture toughness of Pro.3 can not meet technical standard while the lamellar α has become quite thick. In addition to lamellar α thickness, fracture toughness is also influenced by other factors, such as grain boundary width, primary β grain size, etc. [15]. As arrow A in Fig. 9(b) shows, the crack presents a largeangle deflection at the grain boundary, which makes the crack propagation path much more tortuous and crack length increases. Thus the crack propagation rate decreases and fracture toughness improves significantly. The thick grain boundary α layer of Pro.1 can absorb quantities of energy when cracks propagate from one grain to another, and fracture toughness improves [26]. So the thickness of grain boundary also has a significant influence on fracture toughness of BT25y alloy. In addition, with the increase of strain rate, volume fraction of lamellar α and β grain size decrease with only small amplitude. Therefore, it can be concluded that thickness of lamellar α and width of grain boundary α are two main factors influencing the fracture toughness of BT25y alloy. Tensile properties of Pro.1 and Pro.2 after thermal exposure under 550 °C with 100 h are given in Table 4. The tensile strength and yield strength after thermal exposure increase slightly when compared with the RT tensile properties in Table 3. It can be observed from Fig. 10 that the secondary acicular α in Fig. 2 (a) coarsens after long time thermal exposure, and thick strip α appears in local region so that the plastic parameters including elongation and reduction to failure all drop sharply. Large quantities of lamellar α intertwines disorderly, and small β grains grow up during the thermal exposure, which makes discontinuous boundary develop into continuous grain boundary. Thus, the continuous grain boundary turns as weak links during deformation, which reduces the thermal stability especially elongation and reduction to failure. 3.3. Fracture morphology Fig. 11 shows the tensile fracture morphology of specimens forged at different strain rates. For RT tensile deformation, dimples and tear ridges can be observed in all the three processes, but the size of dimples and length of tear ridges are different from each

Fig. 9. Crack propagation path in the specimen after fracture toughness test.

360

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

Table 4 Thermostability of BT25y alloy at different strain rates. Pro. Thermal exposure condition

1 2

550 °C  100 h 550 °C  100 h

UTS

YS

Elongation Reduction to failure sb/MPa ss/MPa δ/% Ψ/% 1215 1186

1122 1106

4.75 9.0

10.0 13.0

depth, and punctual cracks spread outwards from the central of quasi cleavage facets, which can be judged as a typical quasi cleavage fracture. Almost no cleavage facets can be observed in the fracture of Pro.2, and the fracture is full of larger dimples with white secondary phase particles in the bottom, showing that the fracture mode is mainly ductile fracture. The dimples of Pro.3 present a relatively small diameter, but its depth is still much deeper than that of Pro.1. Besides, shorter tear ridges and small fracture facets appear in the fracture, showing that the fracture mode is a mixture of ductile fracture and transgranular cleavage fracture where ductile fracture plays the dominant role. Fig. 11 (d) shows the fracture surface after HT tensile for Pro.2. The fracture surface is full of dimples with large diameter and deep depth, so the HT tensile samples own high tensile plasticity.

4. Conclusions

Fig. 10. The post-exposed microstructure of BT25y alloy with forging strain rate of 1  10  1 s  1.

other. The dimple diameter decreases after a first rise with the decrease of strain rate. There are a lot of large tear ridges in the fracture of Pro.1, the dimples have small diameter and shallow

The effect of isothermal forging strain rate on microstructures and mechanical properties are investigated. The microstructure parameters are measured by the IPP analysis software. Properties like RT tensile deformation, HT tensile deformation, fracture toughness, thermal stability and fracture SEM are performed. Quantitative relationships between microstructure parameters and mechanical properties are obtained. Some main conclusions are as follows: (1) Strain rate has a significant influence on the microstructure and mechanical properties of BT25y alloy during isothermal forging. The microstructure and mechanical properties of BT25y alloy deformed at strain rate of 1  10  2 s  1 can achieve a good match.

Fig. 11. Fracture surfaces of tensile specimens for different processes of BT25y alloy (a) Pro.1 after RT tensile deformation; (b) Pro.2 after RT tensile deformation; (c) Pro.3 after RT tensile deformation; (d) Pro.2 after HT tensile deformation.

X. Yang et al. / Materials Science & Engineering A 673 (2016) 355–361

(2) As strain rate decreases, the lamellar α coarsens, the grain boundary of β grain becomes indistinct, and the microstructure uniformity increases. With the decrease of strain rate, the tensile strength for the three processes all maintains at a high level, but the elongation shows a tendency of first increase and then decrease. Fracture toughness reaches larger values at high strain rate for densely interlaced α colony and thick grain boundary α layer. After long time thermal exposure, the plasticity of 1  10  1 s  1 drops a lot. (3) By observing the SEM fractographs of tensile specimens, it can be concluded that with the decrease of strain rate, the fracture modes correspond to quasi-cleavage fracture, ductile fracture, and a mixture of ductile fracture and brittle fracture, respectively.

Acknowledgment This study was financially supported by the National Natural Science Foundation of China (No. 51205319) and the National Natural Science Foundation of Shaanxi Province (No. 2015JQ5152).

[9]

[10]

[11]

[12] [13] [14] [15]

[16]

[17] [18]

[19]

References [1] E.O. Ezugwu, Z.M. Wang, Titanium alloys and their machinability-a review, J. Mater. Process. Technol. 68 (3) (1997) 262–274. [2] Z.F. Shi, H.Z. Guo, C. Qin, H.Q. Liang, Z.K. Yao, A method to determine main microstructural features influencing mechanical properties of two-phase titanium alloys, Mater. Sci. Eng. A 611 (2014) 136–141. [3] X. Ma, W.D. Zeng, F. Tian, Y.G. Zhou, The kinetics of dynamic globularization during hot working of a two phase titanium alloy with starting lamellar microstructure, Mater. Sci. Eng. A 548 (2012) 6–11. [4] S.Y. Wei, Q.J. Wang, Y. He, Z.Q. Chen, Y.Y. Liu, Review of BT25 and BT25y titanium alloys for aero-engine, Titan. Ind. Prog. 30 (2013) 9–14. [5] Z. Zhang, Y. Yu, S.X. Hui, M.Y. Chu, J.Y. Shen, Research on the ordered phase in BT25y titanium alloy, Rare Metal. Mater. Eng. 37 (2008) 82–88. [6] Z.X. Du, J.S. Liu, S.D. Jiang, S.L. Xiao, F.T. Kong, Y.Y. Chen, Strain rate dependence of microstructural evolution in β titanium alloy during subtransus superplastic deformation, J. Alloy. Compd. 647 (2015) 1–5. [7] T.B. Wang, B.L. Li, M. Li, Y.C. Li, Z.Q. Wang, Z.R. Nie, Effects of strain rates on deformation twinning behavior in α-titanium, Mater. Charact. 106 (2015) 218–225. [8] D. He, J.C. Zhu, S. Zaefferer, D. Raabe, Y. Liu, Z.L. Lai, X.W. Yang, Influences of deformation strain, strain rate and cooling rate on the Burgers orientation

[20] [21]

[22]

[23] [24]

[25]

[26]

361

relationship and variants morphology during β-α phase transformation in a near α titanium alloy, Mater. Sci. Eng. A 549 (2012) 20–29. J. Zhang, Y. Wang, Effect of strain rate on the tension behavior of Ti-6.6Al3.3Mo-1.8Zr-0.29Si alloy at low temperatures, Mater. Sci. Eng. A 605 (2014) 59–64. C. Li, J.H. Chen, X. Wu, S. van der Zwaag, Effect of strain rate on stress-induced martensitic formation and the compressive properties of Ti-V-(Cr, Fe)-Al alloys, Mater. Sci. Eng. A 573 (2013) 111–118. S.Q. Wang, J.H. Liu, D.L. Chen, Effect of strain rate and temperature on strain hardening behavior of a dissimilar joint between Ti-6Al-4V and Ti17 alloys, Mater. Des. 56 (2014) 174–184. H.Z. Guo, Forging of Alloy Steel and Nonferrous Metals, Northwestern Polytechnical University Press, Xi’An, 2009. Z.E. Liu, Material Science and Engineering, Northwestern Polytechnical University Press, Xi’An, 2007. R. Ding, Z.X. Guo, A. Wilson, Microstructural evolution of a Ti-6Al-4V alloy during thermomechanical processing, Mater. Sci. Eng. A 327 (2002) 233–245. T. Wang, H.Z. Guo, Y. Zhao, Z.K. Yao, Y.G. Guo, Z.L. Dou, Effect of isothermal deformation amount on microstructure and properties of 600 °C high temperature titanium alloy TG6, Rare Metal. Mater. Eng. 39 (2010) 1540–1544. C. Sauer, G. Luetjering, Thermo-mechanical processing of high strength β-titanium alloys and effects on microstructure and properties, J. Mater. Process. Technol. 17 (2001) 311–317. C. Sauer, G. Luetjering, Influence of α layers at β grain boundaries on mechanical properties of Ti-alloys, Mater. Sci. Eng. A 319- 321 (2001) 393–397. X.N. Peng, H.Z. Guo, T. Wang, Z.K. Yao, Effects of β treatments on microstructures and mechanical properties of TC4-DT titanium alloy, Mater. Sci. Eng. A 533 (2012) 55–63. T. Wang, H.Z. Guo, Y.W. Wang, X.N. Peng, Y. Zhao, Z.K. Yao, The effect of microstructure on tensile properties, deformation mechanisms and fracture models of TG6 high temperature titanium alloy, Mater. Sci. Eng. A 528 (2011) 2370–2379. L.W. Tsay, C.L. Hsu, C. Chen, Notched tensile fracture of Ti-4.5Al-3V-2Fe-2Mo welds at elevated temperatures, Mater. Chem. Phys. 120 (2010) 715–721. T. Wang, H.Z. Guo, Y.W. Wang, Z.K. Yao, Influence of processing parameters on microstructure and tensile properties of TG6 titanium alloy, Mater. Sci. Eng. A 528 (2010) 736–744. Y.Q. Ning, Z.K. Yao, H. Li, H.Z. Guo, Y. Tao, Y.W. Zhang, High temperature deformation behavior of hot isostatically pressed P/M FGH4096 superalloy, Mater. Sci. Eng. A 527 (2010) 961–966. B.C. Zhang, Non-ferrous Alloy and Heat Treatment, Northwestern Polytechnical University Press, Xi’An, 1993. Y.J. Ma, J.W. Li, J.F. Lei, Z.Y. Tang, Y.Y. Liu, R. Yang, Influences of microstructure on fatigue crack propagating path and crack growth rates in TC4ELI alloy, Acta Metall. Sin. 46 (2010) 1086–1092. T. Wang, H.Z. Guo, Y.Q. Zhang, Z.K. Yao, L.J. Tan, Effects of hot forging temperature on microstructure and mechanical property of TG6 high temperature titanium alloy, Acta Metall. Sin. 46 (2010) 913–919. S. Ankem, H. Margolin, C.A. Greene, B.W. Neuberger, P.G. Oberson, Mechanical properties of alloys consisting of two ductile phases, Prog. Mater. Sci. 51 (2006) 632–709.