- Email: [email protected]
Effect of aging on the tensile properties and microstructures of a near-alpha titanium alloy
Materials and Design 58 (2014) 108–115
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Effect of aging on the tensile properties and microstructures of a near-alpha titanium alloy Weiju Jia a,b,⇑, Weidong Zeng a, Hanqing Yu a a b
State Key Laboratory of Solidification Processing, School of Materials, Northwestern Polytechnical University, Xi’an 710072, China Northwestern Institute for Nonferrous Metal Research, Xi’an 710016, China
a r t i c l e
i n f o
Article history: Received 12 September 2013 Accepted 24 January 2014 Available online 5 February 2014 Keywords: Near-alpha titanium alloy Aging treatment Tensile properties Microstructure
a b s t r a c t The effect of aging temperature between 650 °C and 750 °C for different aging times on the tensile properties and microstructures of Ti60 alloy were studied. The results show that the strength of the alloy increases first and then decreases with the aging temperature increases from 650 °C to 750 °C. The reduction of area of the alloy is more sensitive to the aging time than elongation. With increasing aging temperature and time, the volume fracture and grain size of silicides and a2 phase increase gradually. The silicides have the strengthen effect on the Ti60 alloy, but the effect weakens when the silicides grow up. The loss of ductility is mainly attributed to the precipitation of a2 phase after aging treatment. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction With the requirement of aircraft construction integrity, reliability and durability, near-a titanium alloy which can be serviced at 600 °C, have been explored competitively all over the world [1]. The characteristic of this kind of alloy is the addition of the silicon to improve the high-temperature tensile and creep properties. The superiority of the alloys is most pronounced in the solution-treated and aged (STA) condition with a precipitation hardened microstructure. However, the practical application of these alloys has been limited by their ductility. A few possibilities have been proposed for the loss of ductility in these alloy systems including the precipitation of intermetallic phases (Ti3Al) and silicides, and formation of the oxides on the surface of the alloy [2–4]. Donlon et al. [5] pointed out that the a2 (Ti3Al) precipitates significantly decrease the ductility of titanium alloys, and this effect is further exacerbated by the precipitation of silicides. Cai et al. [6] also found that a2 phase can decrease the ductility of TG6 titanium alloy. Neal and Fox [7] have shown that reduction in ductility of IMI834 alloy is essentially due to silicides. Several investigations have shown that the oxidation layer can modify the mechanical properties of titanium alloy due to the formation of surface cracks [3,6,8]. Meanwhile, a large amount of reports indicated that the mechanical ⇑ Corresponding author at: State Key Laboratory of Solidification Processing, School of Materials, Northwestern Polytechnical University, Xi’an 710072, China. Tel./fax: +86 29 86231078. E-mail addresses: [email protected], [email protected] (W. Jia). http://dx.doi.org/10.1016/j.matdes.2014.01.063 0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
properties of the titanium alloy are strongly influenced by the volume fraction, grain size, morphology and distribution of the precipitation which in turn depends on the composition, hot working processing and heat treatment of the alloy [9,10]. Ti60 alloy, the alloy of present investigation, is a new hightemperature titanium alloy developed in China. It belongs to Ti–Al–Sn–Zr–Mo–Si series titanium alloy, which is similar to IMI834 alloy. Compared with IMI834 alloy, more silicon element is added to improve its creep performance at high temperature. In addition, a small amount of tantalum and carbon elements are added in the alloy to improve its heat-resistant and widen the processing window, respectively [11]. Several investigations have been performed to characterize the constitutive relationship, the hot deformation behavior and the heat treatment of Ti60 alloy [12–14]. Our previous researches found that two instable processing fields should be avoided due to the flow localization during the hot deformation of Ti60 alloy [13]. Cai et al. [15] have researched the effect of solution treatment on the creep properties of Ti60 alloy forged in the (a + b) phase field. The results showed that the excellent creep properties were obtained for the specimens with lamellar structure which obtained by solution treated above the b transus following air cooling. Meanwhile, for the specimen solution treated under the b transus, the primary a phase decreases with increasing of solution temperature which can improve the creep properties of the alloy with equiaxed structure. Hao et al. [16] have researched the influence of aging on the creep properties of Ti60 alloy. They found that the aging temperature of the alloy should be in the range of 740–760 °C to improve its creep properties. The aging time cannot exceed 2 h at 750 °C because the long
109
W. Jia et al. / Materials and Design 58 (2014) 108–115
time aging will results in the precipitation of silicides, and thus the content of Si in the solution state is decreased, then the creep properties is decreased. Wang et al. [17] have investigated the influence of forging processes on microstructure and mechanical properties of Ti60 alloy. The results indicated that the bimodal microstructure was obtained after near-b forging which have excellent properties. The heat treatment variables are time and temperature of solution treatment, cooling rate, and aging temperature and time. The present paper is concerned with the influence of aging temperature and time on the microstructure and properties of Ti60 alloy. This work is part of a wider study being carried out with the aim of developing thermal and thermo-mechanical treatments for optimum mechanical properties in the high temperature titanium alloy Ti60. 2. Experimental details Ti60 alloy used in the present work was received in bar form with a diameter of 270 mm. Its chemical composition is presented in Table 1. The b transus temperature (Tb) of the alloy was identified as 1050 °C using metallographic analysis method by the heating test, in which four specimens with the same size of U10 mm 10 mm were heated at 1040, 1045, 1050 and 1055 °C, held for 40 min and then quenched immediately, respectively, and consequently the volume fraction of a phase of the specimen heated at 1050 °C was nearly disappeared. To obtain the desired microstructure, the as-received Ti60 alloy bar had been subjected to a large number of hot forging which is carried out through open die forging in the a + b region, with sequential deformations in lengthwise and radial directions. This process was repeated for three times, with reheating stages in between. Finally, a homogenous bimodal microstructure was obtained which consisted of equiaxed a phase within a fine transformed b matrix, as shown in Fig. 1. Different heat treatments were given to the material, as shown in Table 2. First, all the specimens were solution treated (ST) at 1010 °C for 2 h following oil cooling (OC), and then the specimens were aged at 650 °C, 700 °C and 750 °C for different times, separately. Cylindrical tensile specimens with a gauge length of 25 mm and a diameter of 5 mm were employed. The tensile tests were conducted per ISO 6892 standard [18]. The properties results were the mean values obtained from three tested specimens. Microstructures of the specimens were analyzed by an optical microscope (OM) Leica DFC320. TEM specimens with a thickness of 0.3 mm were prepared by cutting in the tensile specimens and mechanically thinned to 40–50 lm in thickness. After jet polishing by a double-jet polisher at 30 °C, specimens were observed in HITACHI H-800 transmission electron microscope at 175 kV.
Fig. 1. Initial microstructure of Ti60 alloy.
Table 2 Heat treatment given to Ti60 alloy. Solution treatment (°C)
Aging temperature (°C)
Aging time (h)
Cooling way
1010
650 700
0/2/4/8/16 0/2/4/8/16/24/ 48 0/2/4/8/16
Air cooling
750
Fig. 2. Effect of aging temperature on the tensile properties of Ti60 alloy.
3. Results 3.1. Tensile properties The room-temperature tensile properties of Ti60 alloy after aging at 650 °C, 700 °C and 750 °C for 2 h are shown in Fig. 2. It can be seen that the reduction of area (RA) and elongation (EI) of the specimens have a little change with increasing aging temperature. It is suggested that the ductility of the alloy is not sensitive to the aging temperature under the condition of short aging time. The
Table 1 Chemical composition of Ti60 alloy (wt%). Al
Sn
Zr
Mo
Nb
Ta
Si
C
Ti
5.8
4.0
3.5
0.4
0.4
1.0
0.4
0.06
Balance
yielding strength of the alloy increases when the aging temperature increasing from 650 °C to 700 °C. However, when the aging temperature increases to 750 °C, the yielding strength decreases obviously and the value even lower than that of specimens aged at 650 °C. The results suggested that the change of internal microstructures may make the alloy strengthened with increasing the aging temperature, but the strengthening effect decreases with the aging temperature further increasing. Fig. 3 shows the tensile properties of Ti60 alloy after aging at 650 °C, 700 °C and 750 °C for different times. Compared with the specimens without aging, the strength increases slightly while the ductility decreases obviously for all the specimens after aging for 2 h, especially the decreases of RA is more pronounced than that of elongation. It is suggested that the RA is more sensitive to the aging time than the EI of the alloy. The RA decreases gradually
110
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
Fig. 3. The effect of aging time on the tensile properties of Ti60 alloy aged at (a) 650 °C, (b) 700 °C and (c) 750 °C.
with increasing of aging time at 650 °C, and the steady-state conditions are not observed even aging for 16 h. For the specimens aged at 700 °C, the ductility of the specimens continuous decreases until aging for 16 h. It seems to reach a steady-state condition after aging for 16 h. At 750 °C, the ductility decreases in the beginning but it recovered when the aging time extend to 8 h. It is worth to note that the ductility of the specimens aged at 750 °C is higher than that of specimens aged at 650 °C and 700 °C. 3.2. Microstructure 3.2.1. Effect of aging temperature on the microstructure of the alloy Fig. 4 shows the microstructures of the specimens aged at 650 °C for 2 h. The microstructure consisted of equiaxed a phase within a fine transformed b matrix, as shown in Fig. 4a. TEM images show that a small number of elliptical silicides precipitate from the residual b phase, and the residual b phase is continuous between a platelets, as shown in Fig. 4b. Owing to the small size of the silicide, the diffraction patterns could not be obtained. However, its elliptical morphology may provide a clue that the silicides observed in the alloy may be of the (Ti, Zr)6Si3 type [19]. No silicides precipitate within the a platelets or equiaxed a phase, as shown in Fig. 4c, which suggests the low levels of silicon in the a phase. Selected-area diffraction (SAD) patterns of the primary a phase (ap) exhibit only a reflections indicating that no ordering occurred within the ap or the ordering particles too small to be detected. In addition, a large amount of dislocations can be observed within the a phase, which indicated that the specimen suffered a uniform deformation during the tensile test. Therefore,
the specimen exhibited good plasticity properties, as shown in Fig. 2. When the aging temperature increased to 700 °C, the residual b phase decomposed and more silicides precipitate along the a/b interface. The long axes of the elliptical silicides are seen to be lying nearly parallel to the interfaces of a/b, as shown in Fig. 5a. Meanwhile, silicides are also observed at the boundaries of ap/as, as shown in Fig. 5b. Selected-area diffraction (SAD) patterns of the ap phase indicates that no ordering occurred within the ap yet. Fig. 6 shows the TEM images of specimens aged at 750 °C for 2 h. It can be seen that the residual b phase nearly disappeared and a large amount of silicides precipitate along the a/b interface as well as the boundaries of ap/as (Fig. 6a and b). It is obviously that the size of the silicides increases with increasing temperature. However, no silicides precipitate between the boundaries of colony a with different directions, as shown in Fig. 6a. In addition, a large amount of dislocations are piled up at the grain boundaries of ap, as shown in Fig. 6c. Fig. 6d shows the selected area diffraction pattern from the ap, very weak superlattice diffraction pattern are found obviously which indicates that the ordered a2 phase has precipitated from a matrix. However, owing to the weak diffraction pattern, the dark-field micrograph of a2 phase could not be obtained. Combined with the results of the tensile properties, it may be indicate that the small silicides can strengthen the alloy, however, the strengthen effect will be decreased with the increasing size of silicides. 3.2.2. Effect of aging time on the microstructure of the alloy Fig. 7 shows the TEM images of Ti60 alloy treated at 1010 °C for 2 h and without aging. The ap/ap boundaries are straight and a
111
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(c)
(b)
αs
αp Selected area Silicide
Fig. 4. Microstructures of Ti60 alloy after aged at 650 °C for 2 h (a) the optical image of the alloy, (b) bright-field micrograph showing silicide along a/b platelet boundaries and (c) bright-field micrograph showing the primary a phase and its selected area diffraction.
(a)
(b)
Silicide
Selected area
Fig. 5. TEM of Ti60 alloy after aged at 700 °C for 2 h.
large amount of dislocations pill up at the boundary of ap/ap and a/b. The electron diffraction technique did not reveal the presence of any other phase in this condition. It is suggested that the silicide is the result of aging treatment. The observation is in accordance with the earlier work carried out on the IMI685 alloy [19]. Fig. 8 shows the TEM images of Ti60 alloy aged at 650 °C for 8 h. The quantity of silicides increases after long time aging compared with that aged for 2 h (Fig. 4). However, the size of the silicides has a little change with increasing the aging time. This may be due to the diffusion rate of silicon is too small at the lower aging temperature of 650 °C. SAD patterns of the ap phase exhibit only a reflections indicating that no ordering occurred within the ap. The TEM images of Ti60 alloy aged at 700 °C for 16 h are shown in Fig. 9. It can be found that a large amount of silicides precipitate at the boundaries of a/b and ap/ap after 16 h aging compared with that aged for 2 h (Fig. 5a). And the residual b phase nearly disappeared, as shown in Fig. 9a. It is interesting to note that the long axes of the elliptical silicides no longer parallel to the interfaces of a/b, which have a nearly 45° angle with the a/b boundary. A large amount of dislocations and dislocation wall are observed
within the ap and pill up at the grain boundaries, as shown in Fig. 9b and c. The selected area diffraction pattern from the ap shows a very strong superlattice diffraction pattern which indicates that the ordered a2 phase has precipitated from a matrix. The dark-field micrograph of ap and its selected area diffraction pattern are shown in Fig. 9d, the little bright pot is the a2 particle. Fig. 10 shows the TEM images of Ti60 alloy aged at 700 °C for 48 h, the quality and morphology of silicides similar with that in Fig. 9. The a2 phase also observed within the ap phase, as shown in Fig. 10c. The size of the a2 phase increases compared with that aged for 16 h. Meanwhile, no a2 phase precipitate within the lamellar a phase. The results suggest that the silicides will not sensitive to the aging time when the aging time longer than 16 h at 700 °C. Increasing the aging time lead to grow up of a2 phase at 700 °C, and the larger a2 phase will not lead to the decrease of ductility. Figs. 11 and 12 show the TEM images of Ti60 alloy aged at 750 °C for 8 h and 16 h. The size of the silicides has a little increase compared with that aging at 750 °C for 2 h. A large amount of dislocations are pinned by the a2 phase precipitates within the
112
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
(d)
Dislocation
Dislocation wall
Fig. 6. TEM of Ti60 alloy after aged at 750 °C for 2 h.
(a)
αp
(b)
Dislocations αp
αp
αp
Fig. 7. TEM of Ti60 alloy with solution treatment at 1010 °C (a) the boundary of ap/ap and (b) the morphology of lamellar a.
Fig. 8. The TEM of Ti60 alloy aging at 650 °C for 8 h (a) the macrostructure and (b) the retained b phase and dislocations.
primary a, as shown in Fig. 11b. The dark-field micrograph of primary a phase showing the a2 precipitation and SAD is shown in Fig. 11c. It is clear that the size of a2 and the distance between the a2 particles are larger than that of aged at 700 °C for 16 h and 48 h. After aging at 750 °C for 16 h, the a2 particles grown up further and contacted with each other, as shown in Fig. 12c.
4. Discussion The above analysis indicates that the precipitation characteristics of the second phase are sensitive to the aging temperature and time. Fig. 13 shows the diagrammatic sketch of the precipitation in Ti60 alloy after different aging processing. In the solution
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
(d)
113
αp
Dislocation wall
αp
Fig. 9. TEM of Ti60 alloy aged at 700 °C for16 h (a and b) bright-field micrograph showing silicide along a/b platelet and ap boundaries, (c) bright-field micrograph showing the dislocations in the ap phase and (d) dark-field micrograph of primary alpha phase showing the a2 precipitation and super lattice diffraction spots from a2 phase.
Fig. 10. TEM of Ti60 alloy aging at 700 °C for 48 h (a) bright-field micrograph showing silicide along a/b platelet boundaries, (b) bright-field micrograph showing the dislocations in the ap phase, (c) dark-field micrograph of primary a phase showing the a2 precipitation and (d) super lattice diffraction spots from a2 phase, 4 3 ==½0 1 1 2 . ½5 1 a a2
condition, the silicon dissolved in the b phase and a little dissolved in a phase. These elements were inhibited and cannot precipitate during the following oil cooling. So there was nothing precipitation observed in the alloy. At lower aging temperature (650 °C), the precipitation of silicide was not sensitive to the aging time.
This because the precipitation of silicide is a diffusion process, the diffusion rate is slow at low temperatures which results in the limited precipitation even prolong the aging time to 16 h. At higher aging temperatures, the volume fraction and grain size of silicide increased and two obviously changes of silicide occurred
114
W. Jia et al. / Materials and Design 58 (2014) 108–115
Fig. 11. TEM of Ti60 alloy aged at 750 °C for 8 h (a) bright-field micrograph showing silicide along platelet boundaries, (b) bright-field micrograph and (c) dark-field 4 3 ==½0 1 1 2 . micrograph of primary a phase showing the a2 precipitation and SAD with ½5 1 a a2
Fig. 12. TEM of Ti60 alloy aging at 750 °C for 16 h (a) bright-field micrograph showing silicide along platelet boundaries, (b) bright-field micrograph and (c) dark-field micrograph of primary a phase showing the a2 precipitation and its selected area diffraction.
with increasing aging time. One is the morphology of silicides, which evolution from the short-strip to round shape gradually. Another is the orientation relationship between the silicide and a/b boundaries. This phenomenon was also observed by other researches [20,21]. However, there are no consistent conclusions to explain this phenomenon. It may be related to the prior grow orientation of silicide in the present study, which will be studied in
our further researches. For Ti60 alloy, the precipitation of a2 phase depends on the diffusion of Al element in a phase. The nucleation and growth of a2 phase will depends on the diffusion rate. With the increase of aging temperature, the diffusion rate of Al element increases and the nucleus of the a2 phase decreases which results in the volume fraction of the a2 phase decreases and the a2 size grow up quickly with extension of the aging time.
W. Jia et al. / Materials and Design 58 (2014) 108–115
115
(2) The RA of the alloy continuous decreases with increasing aging time at 650 °C, but the steady-state condition can be reached after aging for 16 h and 8 h at 700 °C and 750 °C respectively. (3) The volume fracture and grain size of silicides and a2 phase increase with increasing aging temperature and time. The silicides have the strengthen effect on the alloy, but the effect weakens when the silicides grow up. The loss of ductility is mainly attributed to the precipitation of a2 phase after aging treatment.
Silicide α2 750oC αp βt
700 oC
References 650 oC
0h
2h
16h
48h
Fig. 13. Diagrammatic sketch of the precipitation in Ti60 alloy after different aging processing.
Several investigations have dealt with the effect of second phase on the mechanical properties of titanium alloys [19,22,23]. The conclusion shows that both the a2 particles and silicides are responsible for the decrease in ductility of the near a alloy. However, it is difficult to judge which one plays the major role. On the whole, there are three stages of precipitation in the present work: a certain amount of silicides with no a2 particles, large silicides with small a2 particles, and large silicides with large a2 particles. In the first stage, the tensile property variation was largely determined by the silicides. These silicides are incoherent with matrix, the dislocations cannot cut across the silicides but interact with them during tensile deformation which leads to the stress concentration at the grain and phase boundaries. The strength effect of silicides weakens when the silicides grow up (Fig. 2). But the silicides have no obvious effect to the ductility of the alloy. In the second stage, the large silicides coexist with a large amount of small a2 particles. These ordered a2 particles are cut by moving dislocations which leads to a local reduction in the resolved shear stress along the active slip planes; meanwhile, the dislocation movement becomes more difficult which is the cause of planar slip in titanium alloys [24]. Therefore, stress concentration generated at the grain boundaries which lead to the increase of strength and decrease of the ductility of the alloy. So, it can be concluded that the loss of ductility of Ti60 alloy is mainly attributed to the precipitation of small a2 phase in this stage. In the last stage, the a2 particles grow up which incoherent with matrix, the dislocations move through the a2 phase by bowing between them (Orowan mechanism). Thus, the deformation becomes more uniform and less stress concentration generated in the alloy. So, the ductility of the alloy increased as shown in Fig. 3c.
5. Conclusions The influence of aging temperature and time on the microstructure and properties of Ti60 alloy has been investigated and the following conclusions can be drawn: (1) The strength of the alloy increases first and then decreases with the aging temperature increases from 650 °C to 750 °C. The reduction of area (RA) of the alloy is more sensitive to the aging time than the elongation.
[1] Leyens C, Peters M. Titanium and titanium alloys. Weinheim: Wiley-VCH; 2003. [2] Srinadh KVS, Singh V. Oxidation behavior of the near alpha titanium alloy IMI834. Bull Mater Sci 2004;27:347–54. [3] Evans RW, Hull RJ, Wilshire B. The effects of alpha-case formation on the creep fracture properties of the high-temperature titanium alloy IMI834. J Mater Process Technol 1996;56:492–501. [4] Guleryuz H, Cimenoglu H. Oxidation of Ti–6Al–4V alloy. J Alloys Compd 2009;472:241–6. [5] Donlon WT, Allison JE, Lasecki JV. The influence of thermal exposure on properties and microstructure of elevated temperature titanium alloys. In: Froes FH, Caplan I, editors. Titanium’92 science and technology TMS. PA: Warrendale; 1993. p. 295–302. [6] Cai JM, Huang X, Cao CX, Ma JM. Thermal stability of TG6 titanium alloy and its partial resumption at high temperature. Rare Met Mater Eng 2010;39:1893–8. [7] Neal DF, Fox SP. The influence of silicides on the properties of near-alpha titanium alloys. In: Froes FH, Caplan I, editors. Titanium’92 science and technology. Warrendale, PA: TEM; 1993. p. 287–94. [8] Garbacz H, Lewandowska M. Microstructural changes during oxidation of titanium alloys. Mater Chem Phys 2003;81:542–7. [9] Cremasco A, Andrade PN, Contieri RJ, Lopes ESN, Afonso CRM, Caram R. Correlations between aging heat treatment, x phase precipitation and mechanical properties of a cast Ti–Nb alloy. Mater Des 2011;32:2387–90. [10] Wang LQ, Lu WJ, Qin JN, Zhang F, Zhang D. Effect of precipitation phase on microstructure and superelasticity of cold-rolled beta titanium alloy during heat treatment. Mater Des 2009;30:3873–8. [11] Zhang SZ, Wang B, Liu ZQ, Gao Y, Yang R. Effect of carbon on microstructures and mechanical properties of Ti-60 high-temperature titanium alloy. Chin J Mater Res 2007;21:433–8. [12] Peng WW, Zeng WD, Wang QJ, Yu HQ. Comparative study on constitutive relationship of as-cast Ti60 titanium alloy during hot deformation based on Arrhenius-type and artificial neural network models. Mater Des 2013;51:95–104. [13] Jia WJ, Zeng WD, Zhou YG, Liu JR, Wang QJ. High-temperature deformation behavior of Ti60 titanium alloy. Mater Sci Eng A 2011;528:4068–74. [14] Jia WJ, Zeng WD, Han YF, Liu JR, Zhou YG, Wang QJ. Prediction of flow stress in isothermal compression of Ti60 alloy using an adaptive network-based fuzzy inference system. Mater Des 2011;32:4676–83. [15] Cai JM, Hao MY, Li XM, Du J, Ma JM, Gao Y, et al. Effect of solution treatment on creep properties of Ti60 high temperature titanium alloy. Acta Metall Sinica 1999;35:202–6. [16] Hao MY, Li XM, Du J, Cai JM, Cao CX. The influence of aging on the creep properties of the high temperature titanium alloy Ti60. Acta Metall Sinica 1999;35:207–10. [17] Wang BS, Jia WJ, Qu WM, Yu HQ, Zhou YG, Zeng WD, et al. Influence of forging processes on microstructure and mechanical properties of Ti60 alloy. Titanium Ind Prog 2011;28:8–11. [18] ISO 6892-1:2009 (E), Metallic materials—tensile testing—part 1: method of test at room temperature. International Organization for Standardization. Published in Switzerland; 2009-08-15. [19] Singh AK, Roy T, Ramachandra C. Microstructural stability on aging of an a + b titanium alloy: Ti–6Al–1.6Zr–3.3Mo–0.30Si. Metall Mater Trans A 1996;27A: 1167–73. [20] Cui WF, Liu CM, Zhou L, Lou GZ. Characteristics of microstructures and secondphase particles in Y-bearing Ti-1100 alloy. Mater Sci Eng A 2002;323:192–7. [21] Na P. The research on the effect of a2 phase critical size in high-temperature titanium alloy. In: School of mechanical engineering. shenyang: Shenyang university; 2007. [22] Li J, Wang L, Qin J, Chen Y, Lu W, Zhang D. The effect of heat treatment on thermal stability of Ti matrix composite. J Alloys Compd 2011;509:52–6. [23] Xin SW, Zhao YQ, Zeng WD. Effect of heat treatment on thermal stability of Ti40 alloy. Trans Nonferrous Met Soc China 2007;17:526–31. [24] Zeng WD, Zhou YG. The influence of microstructure on dwell sensitive fatigue in Ti–6.5Al–3.5Mo–1.5Zr–0.3Si alloy. Mater Sci Eng A 2000;290:33–8.
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Effect of aging on the tensile properties and microstructures of a near-alpha titanium alloy Weiju Jia a,b,⇑, Weidong Zeng a, Hanqing Yu a a b
State Key Laboratory of Solidification Processing, School of Materials, Northwestern Polytechnical University, Xi’an 710072, China Northwestern Institute for Nonferrous Metal Research, Xi’an 710016, China
a r t i c l e
i n f o
Article history: Received 12 September 2013 Accepted 24 January 2014 Available online 5 February 2014 Keywords: Near-alpha titanium alloy Aging treatment Tensile properties Microstructure
a b s t r a c t The effect of aging temperature between 650 °C and 750 °C for different aging times on the tensile properties and microstructures of Ti60 alloy were studied. The results show that the strength of the alloy increases first and then decreases with the aging temperature increases from 650 °C to 750 °C. The reduction of area of the alloy is more sensitive to the aging time than elongation. With increasing aging temperature and time, the volume fracture and grain size of silicides and a2 phase increase gradually. The silicides have the strengthen effect on the Ti60 alloy, but the effect weakens when the silicides grow up. The loss of ductility is mainly attributed to the precipitation of a2 phase after aging treatment. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction With the requirement of aircraft construction integrity, reliability and durability, near-a titanium alloy which can be serviced at 600 °C, have been explored competitively all over the world [1]. The characteristic of this kind of alloy is the addition of the silicon to improve the high-temperature tensile and creep properties. The superiority of the alloys is most pronounced in the solution-treated and aged (STA) condition with a precipitation hardened microstructure. However, the practical application of these alloys has been limited by their ductility. A few possibilities have been proposed for the loss of ductility in these alloy systems including the precipitation of intermetallic phases (Ti3Al) and silicides, and formation of the oxides on the surface of the alloy [2–4]. Donlon et al. [5] pointed out that the a2 (Ti3Al) precipitates significantly decrease the ductility of titanium alloys, and this effect is further exacerbated by the precipitation of silicides. Cai et al. [6] also found that a2 phase can decrease the ductility of TG6 titanium alloy. Neal and Fox [7] have shown that reduction in ductility of IMI834 alloy is essentially due to silicides. Several investigations have shown that the oxidation layer can modify the mechanical properties of titanium alloy due to the formation of surface cracks [3,6,8]. Meanwhile, a large amount of reports indicated that the mechanical ⇑ Corresponding author at: State Key Laboratory of Solidification Processing, School of Materials, Northwestern Polytechnical University, Xi’an 710072, China. Tel./fax: +86 29 86231078. E-mail addresses: [email protected], [email protected] (W. Jia). http://dx.doi.org/10.1016/j.matdes.2014.01.063 0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
properties of the titanium alloy are strongly influenced by the volume fraction, grain size, morphology and distribution of the precipitation which in turn depends on the composition, hot working processing and heat treatment of the alloy [9,10]. Ti60 alloy, the alloy of present investigation, is a new hightemperature titanium alloy developed in China. It belongs to Ti–Al–Sn–Zr–Mo–Si series titanium alloy, which is similar to IMI834 alloy. Compared with IMI834 alloy, more silicon element is added to improve its creep performance at high temperature. In addition, a small amount of tantalum and carbon elements are added in the alloy to improve its heat-resistant and widen the processing window, respectively [11]. Several investigations have been performed to characterize the constitutive relationship, the hot deformation behavior and the heat treatment of Ti60 alloy [12–14]. Our previous researches found that two instable processing fields should be avoided due to the flow localization during the hot deformation of Ti60 alloy [13]. Cai et al. [15] have researched the effect of solution treatment on the creep properties of Ti60 alloy forged in the (a + b) phase field. The results showed that the excellent creep properties were obtained for the specimens with lamellar structure which obtained by solution treated above the b transus following air cooling. Meanwhile, for the specimen solution treated under the b transus, the primary a phase decreases with increasing of solution temperature which can improve the creep properties of the alloy with equiaxed structure. Hao et al. [16] have researched the influence of aging on the creep properties of Ti60 alloy. They found that the aging temperature of the alloy should be in the range of 740–760 °C to improve its creep properties. The aging time cannot exceed 2 h at 750 °C because the long
109
W. Jia et al. / Materials and Design 58 (2014) 108–115
time aging will results in the precipitation of silicides, and thus the content of Si in the solution state is decreased, then the creep properties is decreased. Wang et al. [17] have investigated the influence of forging processes on microstructure and mechanical properties of Ti60 alloy. The results indicated that the bimodal microstructure was obtained after near-b forging which have excellent properties. The heat treatment variables are time and temperature of solution treatment, cooling rate, and aging temperature and time. The present paper is concerned with the influence of aging temperature and time on the microstructure and properties of Ti60 alloy. This work is part of a wider study being carried out with the aim of developing thermal and thermo-mechanical treatments for optimum mechanical properties in the high temperature titanium alloy Ti60. 2. Experimental details Ti60 alloy used in the present work was received in bar form with a diameter of 270 mm. Its chemical composition is presented in Table 1. The b transus temperature (Tb) of the alloy was identified as 1050 °C using metallographic analysis method by the heating test, in which four specimens with the same size of U10 mm 10 mm were heated at 1040, 1045, 1050 and 1055 °C, held for 40 min and then quenched immediately, respectively, and consequently the volume fraction of a phase of the specimen heated at 1050 °C was nearly disappeared. To obtain the desired microstructure, the as-received Ti60 alloy bar had been subjected to a large number of hot forging which is carried out through open die forging in the a + b region, with sequential deformations in lengthwise and radial directions. This process was repeated for three times, with reheating stages in between. Finally, a homogenous bimodal microstructure was obtained which consisted of equiaxed a phase within a fine transformed b matrix, as shown in Fig. 1. Different heat treatments were given to the material, as shown in Table 2. First, all the specimens were solution treated (ST) at 1010 °C for 2 h following oil cooling (OC), and then the specimens were aged at 650 °C, 700 °C and 750 °C for different times, separately. Cylindrical tensile specimens with a gauge length of 25 mm and a diameter of 5 mm were employed. The tensile tests were conducted per ISO 6892 standard [18]. The properties results were the mean values obtained from three tested specimens. Microstructures of the specimens were analyzed by an optical microscope (OM) Leica DFC320. TEM specimens with a thickness of 0.3 mm were prepared by cutting in the tensile specimens and mechanically thinned to 40–50 lm in thickness. After jet polishing by a double-jet polisher at 30 °C, specimens were observed in HITACHI H-800 transmission electron microscope at 175 kV.
Fig. 1. Initial microstructure of Ti60 alloy.
Table 2 Heat treatment given to Ti60 alloy. Solution treatment (°C)
Aging temperature (°C)
Aging time (h)
Cooling way
1010
650 700
0/2/4/8/16 0/2/4/8/16/24/ 48 0/2/4/8/16
Air cooling
750
Fig. 2. Effect of aging temperature on the tensile properties of Ti60 alloy.
3. Results 3.1. Tensile properties The room-temperature tensile properties of Ti60 alloy after aging at 650 °C, 700 °C and 750 °C for 2 h are shown in Fig. 2. It can be seen that the reduction of area (RA) and elongation (EI) of the specimens have a little change with increasing aging temperature. It is suggested that the ductility of the alloy is not sensitive to the aging temperature under the condition of short aging time. The
Table 1 Chemical composition of Ti60 alloy (wt%). Al
Sn
Zr
Mo
Nb
Ta
Si
C
Ti
5.8
4.0
3.5
0.4
0.4
1.0
0.4
0.06
Balance
yielding strength of the alloy increases when the aging temperature increasing from 650 °C to 700 °C. However, when the aging temperature increases to 750 °C, the yielding strength decreases obviously and the value even lower than that of specimens aged at 650 °C. The results suggested that the change of internal microstructures may make the alloy strengthened with increasing the aging temperature, but the strengthening effect decreases with the aging temperature further increasing. Fig. 3 shows the tensile properties of Ti60 alloy after aging at 650 °C, 700 °C and 750 °C for different times. Compared with the specimens without aging, the strength increases slightly while the ductility decreases obviously for all the specimens after aging for 2 h, especially the decreases of RA is more pronounced than that of elongation. It is suggested that the RA is more sensitive to the aging time than the EI of the alloy. The RA decreases gradually
110
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
Fig. 3. The effect of aging time on the tensile properties of Ti60 alloy aged at (a) 650 °C, (b) 700 °C and (c) 750 °C.
with increasing of aging time at 650 °C, and the steady-state conditions are not observed even aging for 16 h. For the specimens aged at 700 °C, the ductility of the specimens continuous decreases until aging for 16 h. It seems to reach a steady-state condition after aging for 16 h. At 750 °C, the ductility decreases in the beginning but it recovered when the aging time extend to 8 h. It is worth to note that the ductility of the specimens aged at 750 °C is higher than that of specimens aged at 650 °C and 700 °C. 3.2. Microstructure 3.2.1. Effect of aging temperature on the microstructure of the alloy Fig. 4 shows the microstructures of the specimens aged at 650 °C for 2 h. The microstructure consisted of equiaxed a phase within a fine transformed b matrix, as shown in Fig. 4a. TEM images show that a small number of elliptical silicides precipitate from the residual b phase, and the residual b phase is continuous between a platelets, as shown in Fig. 4b. Owing to the small size of the silicide, the diffraction patterns could not be obtained. However, its elliptical morphology may provide a clue that the silicides observed in the alloy may be of the (Ti, Zr)6Si3 type [19]. No silicides precipitate within the a platelets or equiaxed a phase, as shown in Fig. 4c, which suggests the low levels of silicon in the a phase. Selected-area diffraction (SAD) patterns of the primary a phase (ap) exhibit only a reflections indicating that no ordering occurred within the ap or the ordering particles too small to be detected. In addition, a large amount of dislocations can be observed within the a phase, which indicated that the specimen suffered a uniform deformation during the tensile test. Therefore,
the specimen exhibited good plasticity properties, as shown in Fig. 2. When the aging temperature increased to 700 °C, the residual b phase decomposed and more silicides precipitate along the a/b interface. The long axes of the elliptical silicides are seen to be lying nearly parallel to the interfaces of a/b, as shown in Fig. 5a. Meanwhile, silicides are also observed at the boundaries of ap/as, as shown in Fig. 5b. Selected-area diffraction (SAD) patterns of the ap phase indicates that no ordering occurred within the ap yet. Fig. 6 shows the TEM images of specimens aged at 750 °C for 2 h. It can be seen that the residual b phase nearly disappeared and a large amount of silicides precipitate along the a/b interface as well as the boundaries of ap/as (Fig. 6a and b). It is obviously that the size of the silicides increases with increasing temperature. However, no silicides precipitate between the boundaries of colony a with different directions, as shown in Fig. 6a. In addition, a large amount of dislocations are piled up at the grain boundaries of ap, as shown in Fig. 6c. Fig. 6d shows the selected area diffraction pattern from the ap, very weak superlattice diffraction pattern are found obviously which indicates that the ordered a2 phase has precipitated from a matrix. However, owing to the weak diffraction pattern, the dark-field micrograph of a2 phase could not be obtained. Combined with the results of the tensile properties, it may be indicate that the small silicides can strengthen the alloy, however, the strengthen effect will be decreased with the increasing size of silicides. 3.2.2. Effect of aging time on the microstructure of the alloy Fig. 7 shows the TEM images of Ti60 alloy treated at 1010 °C for 2 h and without aging. The ap/ap boundaries are straight and a
111
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(c)
(b)
αs
αp Selected area Silicide
Fig. 4. Microstructures of Ti60 alloy after aged at 650 °C for 2 h (a) the optical image of the alloy, (b) bright-field micrograph showing silicide along a/b platelet boundaries and (c) bright-field micrograph showing the primary a phase and its selected area diffraction.
(a)
(b)
Silicide
Selected area
Fig. 5. TEM of Ti60 alloy after aged at 700 °C for 2 h.
large amount of dislocations pill up at the boundary of ap/ap and a/b. The electron diffraction technique did not reveal the presence of any other phase in this condition. It is suggested that the silicide is the result of aging treatment. The observation is in accordance with the earlier work carried out on the IMI685 alloy [19]. Fig. 8 shows the TEM images of Ti60 alloy aged at 650 °C for 8 h. The quantity of silicides increases after long time aging compared with that aged for 2 h (Fig. 4). However, the size of the silicides has a little change with increasing the aging time. This may be due to the diffusion rate of silicon is too small at the lower aging temperature of 650 °C. SAD patterns of the ap phase exhibit only a reflections indicating that no ordering occurred within the ap. The TEM images of Ti60 alloy aged at 700 °C for 16 h are shown in Fig. 9. It can be found that a large amount of silicides precipitate at the boundaries of a/b and ap/ap after 16 h aging compared with that aged for 2 h (Fig. 5a). And the residual b phase nearly disappeared, as shown in Fig. 9a. It is interesting to note that the long axes of the elliptical silicides no longer parallel to the interfaces of a/b, which have a nearly 45° angle with the a/b boundary. A large amount of dislocations and dislocation wall are observed
within the ap and pill up at the grain boundaries, as shown in Fig. 9b and c. The selected area diffraction pattern from the ap shows a very strong superlattice diffraction pattern which indicates that the ordered a2 phase has precipitated from a matrix. The dark-field micrograph of ap and its selected area diffraction pattern are shown in Fig. 9d, the little bright pot is the a2 particle. Fig. 10 shows the TEM images of Ti60 alloy aged at 700 °C for 48 h, the quality and morphology of silicides similar with that in Fig. 9. The a2 phase also observed within the ap phase, as shown in Fig. 10c. The size of the a2 phase increases compared with that aged for 16 h. Meanwhile, no a2 phase precipitate within the lamellar a phase. The results suggest that the silicides will not sensitive to the aging time when the aging time longer than 16 h at 700 °C. Increasing the aging time lead to grow up of a2 phase at 700 °C, and the larger a2 phase will not lead to the decrease of ductility. Figs. 11 and 12 show the TEM images of Ti60 alloy aged at 750 °C for 8 h and 16 h. The size of the silicides has a little increase compared with that aging at 750 °C for 2 h. A large amount of dislocations are pinned by the a2 phase precipitates within the
112
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
(d)
Dislocation
Dislocation wall
Fig. 6. TEM of Ti60 alloy after aged at 750 °C for 2 h.
(a)
αp
(b)
Dislocations αp
αp
αp
Fig. 7. TEM of Ti60 alloy with solution treatment at 1010 °C (a) the boundary of ap/ap and (b) the morphology of lamellar a.
Fig. 8. The TEM of Ti60 alloy aging at 650 °C for 8 h (a) the macrostructure and (b) the retained b phase and dislocations.
primary a, as shown in Fig. 11b. The dark-field micrograph of primary a phase showing the a2 precipitation and SAD is shown in Fig. 11c. It is clear that the size of a2 and the distance between the a2 particles are larger than that of aged at 700 °C for 16 h and 48 h. After aging at 750 °C for 16 h, the a2 particles grown up further and contacted with each other, as shown in Fig. 12c.
4. Discussion The above analysis indicates that the precipitation characteristics of the second phase are sensitive to the aging temperature and time. Fig. 13 shows the diagrammatic sketch of the precipitation in Ti60 alloy after different aging processing. In the solution
W. Jia et al. / Materials and Design 58 (2014) 108–115
(a)
(b)
(c)
(d)
113
αp
Dislocation wall
αp
Fig. 9. TEM of Ti60 alloy aged at 700 °C for16 h (a and b) bright-field micrograph showing silicide along a/b platelet and ap boundaries, (c) bright-field micrograph showing the dislocations in the ap phase and (d) dark-field micrograph of primary alpha phase showing the a2 precipitation and super lattice diffraction spots from a2 phase.
Fig. 10. TEM of Ti60 alloy aging at 700 °C for 48 h (a) bright-field micrograph showing silicide along a/b platelet boundaries, (b) bright-field micrograph showing the dislocations in the ap phase, (c) dark-field micrograph of primary a phase showing the a2 precipitation and (d) super lattice diffraction spots from a2 phase, 4 3 ==½0 1 1 2 . ½5 1 a a2
condition, the silicon dissolved in the b phase and a little dissolved in a phase. These elements were inhibited and cannot precipitate during the following oil cooling. So there was nothing precipitation observed in the alloy. At lower aging temperature (650 °C), the precipitation of silicide was not sensitive to the aging time.
This because the precipitation of silicide is a diffusion process, the diffusion rate is slow at low temperatures which results in the limited precipitation even prolong the aging time to 16 h. At higher aging temperatures, the volume fraction and grain size of silicide increased and two obviously changes of silicide occurred
114
W. Jia et al. / Materials and Design 58 (2014) 108–115
Fig. 11. TEM of Ti60 alloy aged at 750 °C for 8 h (a) bright-field micrograph showing silicide along platelet boundaries, (b) bright-field micrograph and (c) dark-field 4 3 ==½0 1 1 2 . micrograph of primary a phase showing the a2 precipitation and SAD with ½5 1 a a2
Fig. 12. TEM of Ti60 alloy aging at 750 °C for 16 h (a) bright-field micrograph showing silicide along platelet boundaries, (b) bright-field micrograph and (c) dark-field micrograph of primary a phase showing the a2 precipitation and its selected area diffraction.
with increasing aging time. One is the morphology of silicides, which evolution from the short-strip to round shape gradually. Another is the orientation relationship between the silicide and a/b boundaries. This phenomenon was also observed by other researches [20,21]. However, there are no consistent conclusions to explain this phenomenon. It may be related to the prior grow orientation of silicide in the present study, which will be studied in
our further researches. For Ti60 alloy, the precipitation of a2 phase depends on the diffusion of Al element in a phase. The nucleation and growth of a2 phase will depends on the diffusion rate. With the increase of aging temperature, the diffusion rate of Al element increases and the nucleus of the a2 phase decreases which results in the volume fraction of the a2 phase decreases and the a2 size grow up quickly with extension of the aging time.
W. Jia et al. / Materials and Design 58 (2014) 108–115
115
(2) The RA of the alloy continuous decreases with increasing aging time at 650 °C, but the steady-state condition can be reached after aging for 16 h and 8 h at 700 °C and 750 °C respectively. (3) The volume fracture and grain size of silicides and a2 phase increase with increasing aging temperature and time. The silicides have the strengthen effect on the alloy, but the effect weakens when the silicides grow up. The loss of ductility is mainly attributed to the precipitation of a2 phase after aging treatment.
Silicide α2 750oC αp βt
700 oC
References 650 oC
0h
2h
16h
48h
Fig. 13. Diagrammatic sketch of the precipitation in Ti60 alloy after different aging processing.
Several investigations have dealt with the effect of second phase on the mechanical properties of titanium alloys [19,22,23]. The conclusion shows that both the a2 particles and silicides are responsible for the decrease in ductility of the near a alloy. However, it is difficult to judge which one plays the major role. On the whole, there are three stages of precipitation in the present work: a certain amount of silicides with no a2 particles, large silicides with small a2 particles, and large silicides with large a2 particles. In the first stage, the tensile property variation was largely determined by the silicides. These silicides are incoherent with matrix, the dislocations cannot cut across the silicides but interact with them during tensile deformation which leads to the stress concentration at the grain and phase boundaries. The strength effect of silicides weakens when the silicides grow up (Fig. 2). But the silicides have no obvious effect to the ductility of the alloy. In the second stage, the large silicides coexist with a large amount of small a2 particles. These ordered a2 particles are cut by moving dislocations which leads to a local reduction in the resolved shear stress along the active slip planes; meanwhile, the dislocation movement becomes more difficult which is the cause of planar slip in titanium alloys [24]. Therefore, stress concentration generated at the grain boundaries which lead to the increase of strength and decrease of the ductility of the alloy. So, it can be concluded that the loss of ductility of Ti60 alloy is mainly attributed to the precipitation of small a2 phase in this stage. In the last stage, the a2 particles grow up which incoherent with matrix, the dislocations move through the a2 phase by bowing between them (Orowan mechanism). Thus, the deformation becomes more uniform and less stress concentration generated in the alloy. So, the ductility of the alloy increased as shown in Fig. 3c.
5. Conclusions The influence of aging temperature and time on the microstructure and properties of Ti60 alloy has been investigated and the following conclusions can be drawn: (1) The strength of the alloy increases first and then decreases with the aging temperature increases from 650 °C to 750 °C. The reduction of area (RA) of the alloy is more sensitive to the aging time than the elongation.
[1] Leyens C, Peters M. Titanium and titanium alloys. Weinheim: Wiley-VCH; 2003. [2] Srinadh KVS, Singh V. Oxidation behavior of the near alpha titanium alloy IMI834. Bull Mater Sci 2004;27:347–54. [3] Evans RW, Hull RJ, Wilshire B. The effects of alpha-case formation on the creep fracture properties of the high-temperature titanium alloy IMI834. J Mater Process Technol 1996;56:492–501. [4] Guleryuz H, Cimenoglu H. Oxidation of Ti–6Al–4V alloy. J Alloys Compd 2009;472:241–6. [5] Donlon WT, Allison JE, Lasecki JV. The influence of thermal exposure on properties and microstructure of elevated temperature titanium alloys. In: Froes FH, Caplan I, editors. Titanium’92 science and technology TMS. PA: Warrendale; 1993. p. 295–302. [6] Cai JM, Huang X, Cao CX, Ma JM. Thermal stability of TG6 titanium alloy and its partial resumption at high temperature. Rare Met Mater Eng 2010;39:1893–8. [7] Neal DF, Fox SP. The influence of silicides on the properties of near-alpha titanium alloys. In: Froes FH, Caplan I, editors. Titanium’92 science and technology. Warrendale, PA: TEM; 1993. p. 287–94. [8] Garbacz H, Lewandowska M. Microstructural changes during oxidation of titanium alloys. Mater Chem Phys 2003;81:542–7. [9] Cremasco A, Andrade PN, Contieri RJ, Lopes ESN, Afonso CRM, Caram R. Correlations between aging heat treatment, x phase precipitation and mechanical properties of a cast Ti–Nb alloy. Mater Des 2011;32:2387–90. [10] Wang LQ, Lu WJ, Qin JN, Zhang F, Zhang D. Effect of precipitation phase on microstructure and superelasticity of cold-rolled beta titanium alloy during heat treatment. Mater Des 2009;30:3873–8. [11] Zhang SZ, Wang B, Liu ZQ, Gao Y, Yang R. Effect of carbon on microstructures and mechanical properties of Ti-60 high-temperature titanium alloy. Chin J Mater Res 2007;21:433–8. [12] Peng WW, Zeng WD, Wang QJ, Yu HQ. Comparative study on constitutive relationship of as-cast Ti60 titanium alloy during hot deformation based on Arrhenius-type and artificial neural network models. Mater Des 2013;51:95–104. [13] Jia WJ, Zeng WD, Zhou YG, Liu JR, Wang QJ. High-temperature deformation behavior of Ti60 titanium alloy. Mater Sci Eng A 2011;528:4068–74. [14] Jia WJ, Zeng WD, Han YF, Liu JR, Zhou YG, Wang QJ. Prediction of flow stress in isothermal compression of Ti60 alloy using an adaptive network-based fuzzy inference system. Mater Des 2011;32:4676–83. [15] Cai JM, Hao MY, Li XM, Du J, Ma JM, Gao Y, et al. Effect of solution treatment on creep properties of Ti60 high temperature titanium alloy. Acta Metall Sinica 1999;35:202–6. [16] Hao MY, Li XM, Du J, Cai JM, Cao CX. The influence of aging on the creep properties of the high temperature titanium alloy Ti60. Acta Metall Sinica 1999;35:207–10. [17] Wang BS, Jia WJ, Qu WM, Yu HQ, Zhou YG, Zeng WD, et al. Influence of forging processes on microstructure and mechanical properties of Ti60 alloy. Titanium Ind Prog 2011;28:8–11. [18] ISO 6892-1:2009 (E), Metallic materials—tensile testing—part 1: method of test at room temperature. International Organization for Standardization. Published in Switzerland; 2009-08-15. [19] Singh AK, Roy T, Ramachandra C. Microstructural stability on aging of an a + b titanium alloy: Ti–6Al–1.6Zr–3.3Mo–0.30Si. Metall Mater Trans A 1996;27A: 1167–73. [20] Cui WF, Liu CM, Zhou L, Lou GZ. Characteristics of microstructures and secondphase particles in Y-bearing Ti-1100 alloy. Mater Sci Eng A 2002;323:192–7. [21] Na P. The research on the effect of a2 phase critical size in high-temperature titanium alloy. In: School of mechanical engineering. shenyang: Shenyang university; 2007. [22] Li J, Wang L, Qin J, Chen Y, Lu W, Zhang D. The effect of heat treatment on thermal stability of Ti matrix composite. J Alloys Compd 2011;509:52–6. [23] Xin SW, Zhao YQ, Zeng WD. Effect of heat treatment on thermal stability of Ti40 alloy. Trans Nonferrous Met Soc China 2007;17:526–31. [24] Zeng WD, Zhou YG. The influence of microstructure on dwell sensitive fatigue in Ti–6.5Al–3.5Mo–1.5Zr–0.3Si alloy. Mater Sci Eng A 2000;290:33–8.