2019 年 60 巻 9 号 p. 1733-1739
In this study, the microstructure, tensile strength, elongation, and reduction of area of near-β Ti alloys (Ti-17) were investigated after being subjected to solution and aging treatments. Ti-17 was forged at temperatures between 700 and 850°C followed by air cooling. Then, the forged Ti-17 was subjected to solution treatment at 800°C for 4 h followed by water quenching and aging treatment at 620°C for 8 h followed by air cooling. Tensile tests were performed at room temperature, 450°C, and 600°C. The change in microstructure at different forging temperatures was exhibited by only the volume fraction and morphology of the grain boundary (GB) α phase. That is, a granular GB α phase was formed in the samples forged at 700 and 750°C. Moreover, a film-like GB α phase was formed in the samples forged at 800 and 850°C. The tensile strength was the same for all the tested samples, indicating that the microstructure has little effect on the tensile strength. The elongation and reduction of area increased with decreasing volume fraction in the GB α phase. It is considered that the film-like morphology slightly improves ductility.
Fig. 8 Elongation and reduction of area for volume fraction of GB α phase.
The Ti-17 alloy (Ti–5Al–2Sn–2Zr–4Cr–4Mo, mass%) is a near-β (bcc) Ti alloy that consists of α (hcp) precipitates and β matrix. The Ti-17 alloy is stronger than the conventional Ti alloy, Ti–6Al–4V, and is used up to 450°C as a compressor disk in jet engines because of its balanced creep resistance and low cycle fatigue properties.1) Generally, jet engine disks are produced by hot forging in the β phase field or the α and β phase field depending on the desired microstructure, and the subsequent solution and aging treatment. The mechanical properties of Ti alloys are highly influenced by their microstructures, which can be controlled by thermomechanical treatments.
In addition, there are many studies on Ti-17 alloys to understand microstructure formation and the relationship between microstructures and mechanical properties. For example, transformation kinetics, i.e., precipitation of the α phase on the β phase at different cooling rates and isothermal transformation at different temperatures in the α + β phase were investigated, and the prediction model of volume fraction of the α phase was developed using the Johnson–Mehl–Avrami–Kolmogorov model.2) The same group then improved their model to predict α precipitation considering deformation in the β phase.3) Generally, an α phase with acicular, lamellar, or plate-like morphology is obtained from the β phase upon cooling. However, an equiaxed or globularized α microstructure exhibits better balance in terms of strength and ductility compared with acicular or lamellar microstructure; thus, globularization of the α phase has been investigated extensively.4–10) To achieve globularization, isothermal forging was performed in the α + β phase at 820 or 860°C for a fully lamellar microstructure with different height reductions, i.e., 20, 40, 60, and 80%, and subsequent heat treatment for duration ranging from 10 min to 8 h at 820 or 860°C. It was observed that static globularization was controlled by the amount of strain during hot deformation and is not controlled by the deformation temperature.6) The prediction model to estimate the size of the globularized α phase was proposed using the Lifshitz–Slyosov–Wagner theory.7)
As discussed above, the microstructure is strongly controlled by the processing performed on it. To understand the hot deformation behavior, the processing map, according to the dynamic materials model, is useful because it facilitates optimizing conditions for processing. For Ti-17 alloys, several studies have been performed on the processing map and hot deformation behavior at deformation temperatures ranging from 780 to 860°C and strain rates ranging from 10−3 to 10/s.11–14) The prediction model of the stress-strain curve was constructed using regression and an artificial neural network.14) Mechanical properties, such as tensile and fatigue properties, were also investigated, and the relationship between the microstructure and mechanical properties has been discussed.15–17) It was observed that the tensile strength and elongation increased with an increase in the globularization fraction, and the equations to predict the tensile strength and elongation were proposed.15) Furthermore, it was observed that crack propagation in the fatigue test was controlled by the length and thickness of the α platelets. Longer and thicker α platelets contributed to rugged crack propagation, thereby resulting in good plasticity and high fracture toughness.16)
In our previous study, microstructure predictions related to dynamic globularization behavior and change in grain size, and the modeling of a processing map for the forged Ti-17 alloy with a lamellar microstructure were performed using a sample of 5 × 7.5 mm in size.18) The forging temperatures were systematically and widely selected as 750, 800, 850, 900, 950, and 1050°C with strain rates ranging from 10−3 to 1/s. A set of constitutive equations that modeled the microstructural evolution and processing map were established by optimizing the experimental data. These equations were used for FEM analysis in the DEFORM-3D software package, and distributions of the effective strain, effective strain rate, and microstructure factors, such as previous β grain size, aspect ratio of α platelet, and fraction of dynamic globularization of α, after deformation were presented in the 3D images.
Several studies have been performed on the mechanical properties of Ti-17 alloys; however, these studies focused on microstructures controlled only by heat treatment. In addition, the microstructures were influenced by the whole processing stage. Therefore, it is important to understand the effect of both the forging condition and heat treatment condition. In this study, to understand the correlations between the forging condition, microstructure, and tensile properties of Ti-17 alloys, the samples forged in temperatures ranging from 700 to 850°C were prepared. The evolution of microstructures at different forging temperatures with subsequent heat treatment and tensile properties were investigated. The correlations between the volume fraction, size of the α phase, tensile strength, and elongation were discussed.
Cylindrical Ti-17 ingots with a diameter of 134 mm and a height of 192 mm were isothermally forged at 700, 750, 800, and 850°C with a strain rate of 0.033/s using a 1500 t forging press, and then air cooled. The final size of the forged materials was approximately 270 mm in diameter and 50 mm in height. The forged pancake-like materials were solution-treated at 800°C for 4 h followed by water quenching, and aged at 620°C for 8 h followed by air cooling.
The tensile test specimens with 4 mm diameter and 16 mm gage length were cut parallel to the direction of circumference from the D/4 position (D is diameter) in the heat-treated materials. The tensile test was performed according to the ASTM E8 condition at room temperature (RT) and ASTM E21 condition at 450 and 600°C.
The microstructure was observed as a sample near the tensile test specimen at the D/4 position using scanning electron microscopy (SEM, JEOL 7000F) with electron backscatter diffraction (EBSD) at 20 kV. For reference, the microstructure of as-forged samples at the D/4 position was also observed. The volume fraction and the size of primary α and GB α phases were identified using EBSD and SEM image analysis by OLYMPUS Stream software, respectively. The total volume fraction of the α phase was measured using X-ray diffractometry (Bruker, Advance D8). Further, the volume fraction of secondary α was calculated by subtracting the volume fraction of primary α and GB α from the total volume fraction of the α phase because it is difficult to identify the volume fraction of secondary α by EBSD because of the small size.
X-ray measurement was performed on a plate with dimensions of 5 × 10 × 1 mm at RT using Co Kα radiation operated at 40 kV and 35 mA. To remove the effect of texture, the samples were rotated and tilted between 10 and 50° during measurement. The volume fraction was estimated from the ratio between the ideal and experimental intensities of the α and β phases using the following equations:
| \begin{equation} V_{\alpha} = \cfrac{\overline{\displaystyle\sum \cfrac{I^{\alpha}}{R^{\alpha}}}}{\overline{\displaystyle\sum \cfrac{I^{\alpha}}{R^{\alpha}}} + \overline{\displaystyle\sum \cfrac{I^{\beta}}{R^{\beta}}}}, \end{equation} | (1) |
| \begin{equation} V_{\beta} = \cfrac{\overline{\displaystyle\sum \cfrac{I^{\beta}}{R^{\beta}}}}{\overline{\displaystyle\sum \cfrac{I^{\alpha}}{R^{\alpha}}} + \overline{\displaystyle\sum \cfrac{I^{\beta}}{R^{\beta}}}}, \end{equation} | (2) |
The following are the analyzed peaks of the α and β phases:
The microstructures, phase maps analyzed by EBSD, and X-ray diffraction patterns of the forged and solution-treated and aged treated (STA) samples are shown in Figs. 1, 2, and 3, respectively. As shown in Fig. 3, the X-ray diffraction patterns clearly indicate that all samples have the fcc-α and bcc-β phases. As shown in Fig. 2, the phase maps indicate that the plate-like phase with black contrast is identified as the α phase and the matrix is the β phase. In addition to the plate-like α phase formed in grains, different morphologies of the α phase were formed around grain boundaries (GBs), as shown in Fig. 1. In the samples forged at 700 and 750°C, the granular α phase formed in the area around the GBs, as shown in Figs. 1(a) and 1(b). Moreover, a film-like α phase formed at the GBs in the samples forged at 800 and 850°C, as shown in Figs. 1(c) and 1(d). The microstructure difference between the samples forged at different temperatures was highlighted only by the morphology of the GB α phase. In the microstructure with high magnification, a fine α phase was observed in the β matrix as shown in Fig. 4. The plate-like α phase in Fig. 1 is called the primary α phase because it was formed during the solution treatment. The fine α phase in Fig. 4 is called the secondary α phase because it was formed during the aging treatment.
Backscattered electron images of Ti-17 forged samples at (a) 700, (b) 750, (c) 800, and (d) 850°C followed by air cooling subjected to solution treatment at 800°C for 4 h followed by water quenching and aging treatment at 620°C for 8 h followed by air cooling.
Phase map analyzed by EBSD of Ti-17 forged samples at (a) 700, (b) 750, (c) 800, and (d) 850°C followed by air cooling subjected to solution treatment at 800°C for 4 h followed by water quenching and aging treatment at 620°C for 8 h followed by air cooling. The black and gray contrast phases are α and β phases, respectively.
X-ray diffraction patterns of Ti-17 forged samples at (a) 700, (b) 750, (c) 800, and (d) 850°C followed by air cooling subjected to solution treatment at 800°C for 4 h followed by water quenching and aging treatment at 620°C for 8 h followed by air cooling.
High magnification backscattered image of Ti-17 forged samples at 800°C followed by air cooling subjected to solution treatment at 800°C for 4 h followed by water quenching and aging treatment at 620°C for 8 h followed by air cooling.
To understand microstructure evolution, the microstructures of the as-forged samples followed by air cooling were also observed. Figure 5 represents the microstructures of the as-forged samples at (a) 700 and (b) 850°C, respectively. The black dots in Fig. 5(a) are scratches caused by polishing. Figure 5(c) indicates an enlarged microstructure inside a grain of the as-forged sample at 700°C. The primary α phase was already formed in as-forged samples, which is clearly shown in the enlarged image in Fig. 5(c). The GB α phase was also observed as shown in Figs. 5(a) and 5(b). The needle-like α phase with a length of a few µm was formed from GBs to the inside of grains in the as-forged sample at 700°C, as shown in Fig. 5(a). Moreover, a thin film-like α phase was formed at GBs in the as-forged sample at 850°C in Fig. 5(b). The primary α phase was not deformed in the forged sample; thus, it is considered that the primary α phase in grains was not formed during forging but during the air cooling process. A similar observation was made in Ref. 2) and 3). In Ref. 2) and 3), samples of the Ti-17 alloy with length of 30 mm and diameter of 3 mm were cooled at rates between 0.02 and 1°C/s from the β phase field; the α phase formed only around the GBs and no α phase precipitated inside the grains at a cooling rate of 1°C/s. However, the platelet α phase formed inside grains when the cooling rate was slow at 0.75°C/s. This indicates that the α phase forms when the cooling rate is slow and below 0.75°C/s. In our case, the forged sample was very big, with diameter of 270 mm and thickness of 50 mm; thus, the cooling rate at D/4 is slower than 0.75°C/s. Therefore, there is enough time to form the α phase during air cooling.
Backscattered images of the as-forged samples at (a, c) 700 and (b) 850°C, followed by air cooling. (c) Enlarged image of (a).
Unlike the α phase inside grains, the different morphology of GB α phase indicates that the GB α phase was formed during forging at different temperatures. In Ref. 2) and 3), α phase precipitation was observed by isothermal treatment after heat treatment in β phase, followed by cooling to the isothermal treatment temperature. Precipitation of α phase was observed only around GBs for short isothermal treatment. This indicates that GB α phase formation occurred during forging in our study. Furthermore, it is considered that the needle-like α phase at the GB formed during forging grew and changed to a granular shape during solution treatment because of the driving force of deformation strain introduced during forging. Moreover, the film-like α phase at the GB increased in thickness and changed to a thicker film-like α phase during the solution treatment.
The secondary α phase shown in Fig. 4 was not observed in as-forged samples, as shown in Fig. 5(c). It indicates that the secondary α phase was formed during aging. This shows that the forging temperature affected only the morphology of the GB α phase; i.e., the granular shape of the α phase was observed along the GBs for the as-forged samples at 700 and 750°C as shown in Figs. 1(a), 1(b) and 5(a). Moreover, a film-like thin α phase was formed for the as-forged samples at 800 and 850°C as shown in Figs. 1(c), 1(d) and 5(b).
The volume fractions of the primary, secondary, and GB α phases obtained in forged and STA samples were estimated using X-ray diffraction, EBSD analysis, and SEM image analysis by ORLYMPUS Stream software. The results are summarized in Table 1 and plotted by open symbols in Fig. 6. The total volume fraction of the α phase obtained by X-ray diffraction was above 70% and has a large error bar; however, the volume fraction of the equilibrium α phase of Ti-17 was reported as 67%.2,3) Thus, the total volume fraction of the α phase is considered the same in the presently tested samples, and the volume fraction of the α phase obtained by X-ray diffraction was estimated to be higher than that of the equilibrium α phase reported previously.2,3) Further, the volume fraction of the secondary α phase was calculated as the total volume fraction, i.e., 67%. The volume fraction of the primary α phase obtained by EBSD analysis was dispersed and no tendency was observed for the forging temperature. It is caused by microstructure dispersion of the observed area. Thus, the volume fraction of the primary α phase is considered to be the same at approximately 50% in the presently tested samples because of the same solution treatment temperature. Moreover, the volume fraction of the GB α phase exhibited a decreasing trend with the increase in forging temperature. Only GB α phase was affected by the forging temperature, as shown in Fig. 5; thus, it is natural that the volume fraction of the GB α phase decreased with the increasing in forging temperature. Considering the above discussion, the volume fraction of the secondary α phase was re-estimated using total volume fraction of 67%, primary α phase volume fraction of 51%, and measured GB α phase volume fraction. The results are summarized in Table 2 and plotted by the solid symbols in Fig. 6. The results show that the volume fraction of the secondary α phase increased slightly with the rise in the forging temperature. This observation suggests that the microstructure was restored by solution treatment. The size of the primary and GB α phases were also measured. The thickness of the primary α phase was approximately 5.6 µm. The thickness of the GB α phase in the samples forged at 800 and 850°C was approximately 1 µm. The diameter of the GB α phase in the samples forged at 700 and 750°C was 1 µm.
Tensile tests were performed on the forged and STA samples. As shown in Fig. 7(a), the tensile strength and 0.2% proof stress are plotted for different forging temperatures. Both tensile strength and 0.2% proof stress were not affected by the forging temperature and were almost constant at different forging temperatures. The tensile strength and 0.2% proof stress at RT were approximately 1200 and 1100 MPa, respectively, which is consistent with the values for Ti-17 alloys with the similar platelet α phase structure in other studies.16,17) As shown in Fig. 6, the microstructure was restored by solution treatment and the only difference is the volume fraction between the GB and secondary α phases. However, the volume fraction difference between the GB and secondary α phases was very small, within 5%. The constant tensile strength and 0.2% proof stress are controlled by mainly the primary α phase, which is constant and independent of forging temperature, rather than the GB and secondary α phases, which have small difference depending on forging temperature. When the aspect ratio of the primary α phase is large, the alloy possesses high strength because of the interfacial strengthening effect.16) In another reference, the strength has a linear relationship with the volume fraction of the globularized α phase, which can be expressed as a linear equation of the volume fraction of globularized α phase.15) However, in our results, the aspect ratio of α phase was the same for all samples and globularization did not occur in grains. Hence, both the explanations do not reflect our result.
(a) Tensile strength (TS) and 0.2% proof stress (PS) and (b) elongation (EL) and reduction of area (RA) at testing temperature of room temperature, 450, and 600°C.
The elongation and reduction of area exhibited a weak correlation with forging temperature, especially below 450°C. The collected data were varied; however, the elongation and reduction of area decreased with decreasing forging temperature. At 600°C, the reduction of area was almost 100% and the elongation was approximately 50%. The elongation of the sample forged at 700°C was small at 17%; this is because the sample was fractured outside the gage length. The forging temperature changed the volume fraction and morphology of the GB α phase; thus, a correlation exists between the volume fraction of the GB α phase and the elongation, and the reduction of area is indicated in Fig. 8. Below 450°C, it was found that when the volume fraction of the GB α phase is small and the morphology is film-like, the elongation and the reduction of area were slightly increased.
Elongation and reduction of area for volume fraction of GB α phase.
To understand the effect of the GB α phase on crack propagation, the fracture surface of the fractured tensile test specimen was observed. The fracture surface of the sample forged at 700°C is shown in Fig. 9. A drastic increase in reduction of area was observed with the increase in testing temperature in Figs. 9(a), 9(b), and 9(c). The fracture surface of the sample tested at RT was tilted approximately 45° from the tensile direction and the surface was flat, thereby indicating that a fracture occurred by shear deformation as shown in Fig. 9(a). At 450 and 600°C, breaking at the center of the samples was observed as shown in Figs. 9(b) and 9(c). The enlarged fracture surface was almost the same and a ductile dimple pattern was observed for the samples deformed at RT and 450°C as shown in Figs. 9(d) and 9(e). At 600°C, the center part was broken by cavity growth, as shown in Fig. 9(f). The fracture surface of the samples forged at other temperatures possessed the same morphology as in Fig. 9.
Fracture surface of the sample forged at 700°C. Testing temperatures at (a, d) room temperature, (b, e) 450, (c, f) 600°C, respectively. (d), (e), and (f) Enlarged fracture surface of (a), (b), and (c), respectively.
The cross-section near the fracture surface perpendicular to the fracture surface was also observed in Fig. 10. Fractures occurred regardless of the microstructure and cracks propagated through the α and β phases as well as the GB with the GB α phase. Thus, the sample forged at 750°C in Fig. 10(b) indicates that the cracks propagated straight from one grain to the neighboring grain through the granular GB α phase. The sample forged at 700°C indicates that globularization occurred inside the grains, as shown in Fig. 10(a). In the samples forged at 800 and 850°C, a film-like GB α phase was formed and globularization or fragmentation of the film-like GB α phase was observed in the sample forged at 800°C, as shown in Fig. 10(c). It was observed that cracks propagated straight through both of the granular and film-like GB α phases. This observation does not show a clear difference in fracture behavior between the granular and film-like GB α phases. Therefore, further investigation is required to understand tensile plasticity.
The cross-section near the fracture surface perpendicular to the fracture surface of the tested samples at room temperature. Forging temperatures are (a) 700, (b) 750, (c) 800 and (d) 850°C, respectively.
In this study, a near-β Ti alloy, Ti-17, was forged at different temperatures between 700 and 850°C followed by air cooling. Then, these were solution-treated at 800°C for 4 h followed by water quenching and aged at 620°C for 8 h followed by air cooling. The microstructure was investigated in the as-forged samples and the forged and STA samples. The tensile strength, elongation, and reduction of area of the forged and STA samples were investigated at RT, 450°C, and 600°C.
This work was partly supported by Structural Materials for Innovation, Cross-ministerial Strategic Innovation Promotion Program, Cabinet Office, Government of Japan. We are grateful to Ms. H. Gao and T. Bolotova for performing microstructure observation and analysis by SEM-EBSD and SEM image analysis by OLYMPUS Stream software.
