Applicability of As-Cast on β Type Titanium Alloys Proposed in the Compositional Region with Different Tensile Deformation Types

Abstract

Three β type titanium alloys were proposed for as-cast applications in practical fields on the basis of the d-electrons parameters with both bond order (Bo_t) and d-orbital energy level (Md_t). Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe with the lowest Md_t, Ti–4.5Cr–2.5Mn–1.1Al with the lowest Bo_t and Ti–10.8Mo–2.3Sn–1.0Al with the highest Md_t were designed by using ubiquitous elements in the predicted regions showing the slip, twin and martensite dominant deformation behaviors in the Bo_t-Md_t diagram, respectively. Their ingots were produced by the cold crucible levitation melting technique. Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe showed mono β phase and similar stress-strain curves with highest tensile strength more than 1000 MPa at both as-cast and solution treated conditions, which corresponded to the slip dominant deformation. Ti–4.5Cr–2.5Mn–1.1Al showed β and a small amount of α′′ phases, and the stress-strain curves with stress-induced α′′ martensite at both conditions, which corresponded to the twin dominant deformation. Ti–10.8Mo–2.3Sn–1.0Al consisted of β and large amounts of α′′ martensite phases and showed the fracture strain more than 35% at both conditions, which corresponded to the martensite dominant deformation. Segregation degree in solidification process showed 4.8 times larger in the Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe position far from pure Ti position in the Bo_t-Md_t diagram, compared with that of Ti–10.8Mo–2.3Sn–1.0Al close to the pure Ti position. It was found that as-cast application possibility of both alloys of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al could be promising in the view of tension behaviors.

1. Introduction

β type titanium (β-Ti) alloys have been widely investigated for their unique mechanical properties at ambient temperatures, excellent corrosion resistance, shape memory ability and high specific strength.^1–4) Thus, β-Ti alloys are extensively used in aerospace, implant and industrial applications.^5–7) It is known that Ti ranks as the ninth most plentiful element and the fourth most abundant structural metal in the Earth’s crust.⁸⁾ However, remaining drawbacks for Ti alloys is the higher cost than many other light metals because of the long design cycle time, poor refining efficiency, difficulty of melting and complexity of post treatment process etc.⁹⁾ Whereas the cost is the decisive factor for industrial manufacturers when they choose the materials.¹⁰⁾

To reduce the cost of Ti alloys, some measurements are taken by researchers as follows. In the past, the alloy design and development of Ti alloys mainly relied on the repeated experiments and some empirical rules, which are high cost and inefficient, until the propose of d-electrons concept based on the theoretical calculation of electronic structures of alloys. The d-electrons concept was first published by Morinaga et al.^11–13) and has been accepted by numerous Ti alloy researchers.^14,15) Two calculated parameters were mainly utilized in the d-electrons concept. The one is the d-orbital energy level (Md) of alloying transition elements, and the other is the bond order (Bo) that is a measure of the covalent bond strength between atoms. Moreover, Ti is very chemically reactive and easy to react with many elements like oxygen and hydrogen during the manufacturing process, which can greatly influence the properties of Ti alloys.^16–20) The source of these elements usually derives from the gas environment of furnace and the contaminations of the crucible. Therefore, it is essential to control the vacuum level in furnaces and prevent impurities from contaminating. The cold crucible levitation melting (CCLM) method can suit these requests very well. The CCLM consists of a high frequency induction furnace and two electric coils. The upper electric coil is used for heating and the lower electric coil is utilized for levitation the molten metals.^21,22) The molten material can be levitated by the eddy current in the melting crucible so that the alloys can be melted without any contact between materials and the crucible. Moreover, a uniform composition of Ti alloys can be obtained by the diffusion mixing effect and strong stirring from the electromagnetic force.²³⁾

According to the previous reports and practical applications, many complex post-processing treatments were needed after casting for producing conventional β-Ti alloys with high strength and toughness,²⁴⁾ which further led to their high final costs. For example, one of the most widely applied β-Ti alloy Ti–15V–3Cr–3Sn–3Al with the ultimate tensile strength (σ_UTS) of 800∼1400 MPa and fracture strain (ε_f) of 6∼11% can be obtained after solution treated, aged, hot and cold working process.^5,25) The complex post-processing treatments will lead to the high cost and high energy consumption. Herein, when considering the final cost of the Ti alloy for aircraft applications, if the raw material cost is set as 1, the casting cost is about 0.5, the post-processing cost is about 1.5 and the secondary processing cost can be estimated to be 2.²⁶⁾ The post-processing treatment cost is remarkably expensive compared with that of casting, because of the high energy consumption. The development of Ti alloys could be used at as-cast condition is urgent needed to reduce the post-processing and simplified manufacturing process. However, there are almost no precedent for Ti alloys with high alloy using at as-cast condition manufactured just by single casting process. Because the solidification segregation of alloying elements, emergence of non-equilibrium phase and instability of the constituent phases usually appeared in the casting Ti alloys.^27,28) In contrast, it is also reported that many light metals are developed to be applied at as-cast condition which may be an effective way to reduce energy consumption and production costs by omitting complex post-processing procedures, such as aluminum and magnesium alloys.^29,30)

In view of the discussion above, the new composition of three β-Ti alloys were proposed to develop β-Ti alloys for as-cast application. In the present study, according to the d-electrons concept, the new three alloys were chosen by using mainly ubiquitous elements in the predicted regions showing the slip, twin and martensite dominant deformation behaviors in wide range β-Ti alloys area in the Bo_t-Md_t diagram, respectively. The segregation of solute elements was major problem for as-cast application possibility of proposed alloys. The tensile and microstructural properties had to be also investigated at both as-cast and solution treated conditions. Moreover, the variant properties of three β-Ti alloys between as-cast and solution treated conditions as well as their as-cast application possibility were discussed in the present study.

2. Alloy Design for the β Alloys Located on Different Regions in the Bo_t-Md_t Diagram

The Bo_t and Md_t values of three alloys were calculated by utilization of the discrete-variational Xα (DV-Xα) cluster calculation method.^31,32) The parameter values of each element were calculated on the MTi14 cluster model in the case of bcc Ti.²³⁾ The mean values of Bo_t and Md_t were calculated from the following eqs. (1) and (2).   

\begin{equation} \mathit{Bo}_{t} = \sum \mathrm{X}_{\text{i}}(\mathit{Bo})_{\text{i}} \end{equation} (1)

  

\begin{equation} \mathit{Md}_{t} = \sum \mathrm{X}_{\text{i}}(\mathit{Md})_{\text{i}} \end{equation} (2)

Where X_i is the molar fraction of component i in the alloy, (Md)_i and (Bo)_i were the numerical values of Bo_t and Md_t for each component i, respectively.³³⁾ The compositions of three β-Ti alloys in both mol% and mass% as well as their respective Bo_t and Md_t values were showed in Table 1 in detail. The compositional locations of three β-Ti alloys were indicated in Fig. 1. The boundaries of α, α + β and β phases as well as different deformation types (DTs) were also showed in Fig. 1, they were identified by observing the phase types and deformation behaviors in the dozens of commercial and reported alloys.^11,12,34,35) The vectors pointing from pure Ti represent the location of Ti–2 mass%M (M is alloying element selected in this study). This different DT in β-Ti alloys were determined by examining deformation bands which appeared around their Vickers indentations.¹¹⁾

Table 1 The compositions of proposed three β-Ti alloys in both mol% and mass% and their Bo_t and Md_t values.

Fig. 1

The positions indicating chemical compositional of three experimental alloys and their compositional scatters at as-cast and solution treated conditions in the Md_t-Bo_t diagram showing phase and deformation type boundaries. The vectors pointing from pure Ti represent the location of Ti–2 mass%M. (M is alloying element selected in this study)

The new Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe in mass% with the lowest Md_t or high alloy, Ti–4.5Cr–2.5Mn–1.1Al with the lowest Bo_t and Ti–10.8Mo–2.3Sn–1.0Al with the highest Md_t or low alloy, were chosen by using mainly ubiquitous elements in the predicted regions showing the slip, twin and martensite DTs in wide range β-Ti alloys area in the Bo_t-Md_t diagram, respectively. The Bo_t and Md_t values of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy were 2.791 and 2.311, which were approximately same as the Bo_t value of 2.790 of pure titanium and the Md_t value of 2.321 of Ti–15V–3Cr–3Sn–3Al commercial alloy,³⁵⁾ respectively. The Bo_t and Md_t values of Ti–4.5Cr–2.5Mn–1.1Al alloy were 2.781 and 2.375, respectively. They were determined approximately same as 2.784 and 2.375 of Ti–8Cr–6Sn–5Zr commercial alloy.³⁶⁾ The Bo_t value of 2.794 of Ti–10.8Mo–2.3Sn–1.0Al alloy was also approximately same with that of pure Ti. Moreover, the Md_t value of 2.411 of Ti–10.8Mo–2.3Sn–1.0Al alloy was the median value between Ti–4.5Cr–2.5Mn–1.1Al alloy of 2.375 and pure Ti of 2.447. The consideration of solute-segregation in solidification was essential which might cause a position shift for alloys in the Bo_t-Md_t diagram, especially for the multi-element alloys.

3. Experimental Procedures

3.1 Materials and manufacturing process

For the present study, the purity of raw materials for preparing three β-Ti alloys were Ti, Cr, Mn, Fe, Al, Zr, Mo and Sn with 99.8, 99.9, 99.9, 99.9, 99.9, 98.0, 99.0 and 98.5 mass%, respectively. The chemical compositions of alloys were given in mass% units. All ingots were prepared through CCLM (Fuji-CCLM_Fe 1 Kg/100 + 60 kW, Japan) with a water-cooled copper crucible, and the atmosphere of CCLM was argon gas with purity of 99.99% after the pressure less than 3 × 10⁻³ Pa, as seen in Fig. 2. The maximum temperature and holding temperature of CCLM are depending on the melting point of alloying elements. To completely melt the raw materials and well mix of molten metals, the alloys were held for 300 s at maximum temperature of 2300 K in melting process.^4,32) The detailed profiles of temperature in molten metal, electric power in upper and lower coils and pressure in atmosphere of levitation melting and cooling process were given in Fig. 2. Molten metals were solidified in the copper melting crucible after switching off electric power after the melting process. The as-cast specimens were cut directly from the ingot. Whereas, the solution treated specimens were encapsulated in a quartz tube filled with an inert argon atmosphere and held at 1173 K for 3.6 ks. Thereafter, solution treated specimens were water-quenched by breaking the quartz tube and dropped the specimens into cold water.

Fig. 2

Profiles of (a) temperature in molten metal, (b) electric power in upper and lower coils and (c) pressure in atmosphere of levitation melting and cooling process. Abscissa and ordinate are represented with arbitrary scales.

3.2 Microstructural characterization and point analysis

The specimens were mechanically polished by using different grades of polishing paper and etched by using a mixed solution of distilled water, nitric acid and hydrofluoric acid (95:3:2 in volume ratio) and microstructural observation of specimens at as-cast and solution treated conditions were carried out by using an optical microscope (OM). The constituent phases of alloys were determined by X-ray diffraction (XRD) with Cu Kα radiation generated at 40 kV and 30 mA at the room temperature. The chemical compositions of three experimental alloys were measured by point analysis via the EPMA (JEOL JXA-8200, Japan). In order to determine segregation conditions, the compositional scatters through three alloy specimens at as-cast and solution treated conditions were detected with the electron beam of 5 µm in diameter by measuring of about 60 to 80 points. The points analysis thoroughly crossed two grains were measured with the interval of 15 µm between each two points. Moreover, the microstructures of plastic deformed Ti–4.5Cr–2.5Mn–1.1Al alloy at solution treated condition were observed by transmission electron microscopy (TEM). The specimens for TEM were prepared via electrolytic polishing in the solution containing 10% perchloric acid and 90% methanol at 18 V and 288 K (JEM-2000EX II, Japan).

3.3 Tensile test

The tensile specimens with a gauge size of 6 mm in diameter and 20 mm in length, and grips being 10 mm in diameter on both ends were applied in this study. The conditions for tensile test were with an initial strain rate of 1.9 × 10⁻⁴ s⁻¹ at the room temperature. And an extensometer was used for all the tensile testing. The displacement and the constant strain rate were measured by Shimadzu SG 10–50 extensometer during the tensile process, and the stress-strain curve was got by using a mechanical testing machine (Autograph DCS-R-5000, Shimadzu Corporation, Japan). The Vickers hardness measurements were also tested with load of 300 N or 5 N for 10 s (FV-110, Japan).

4. Results and Discussion

4.1 Microstructures and solute segregation in solidification

OM images of three β-Ti alloys were presented in Fig. 3, and their constructed phases were identified using XRD profiles as shown in Fig. 4. For Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy as shown in Fig. 3(a) and (d), there was no significant difference in the microstructures, regardless of heat treatments. The average equiaxed mono β grain sizes of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy at as-cast and solution treated conditions were 350 and 390 µm, respectively. The mixed structure consisting of the β and orthorhombic α′′ martensite phases precipitated in or near the grain boundaries were observed in the as-cast Ti–4.5Cr–2.5Mn–1.1Al alloy as shown in Fig. 3(b). The content of α′′ martensite phase was decreased by solution treatment, as seem in Fig. 3(e). Their average grain sizes were 400 and 450 µm, at as cast and after solution treatment, respectively. In contrast, for Ti–10.8Mo–2.3Sn–1.0Al alloy, the α′′ martensite phase observed at both as-cast and solution treated conditions were shown in Fig. 3(c) and (f), which was consistent with its location showing martensite dominant deformation in the Bo_t-Md_t diagram. Their average grain sizes were 420 and 460 µm, at as-cast and solution treated conditions, respectively. Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al had the Md_t values far and near that of pure Ti in the Bo_t-Md_t diagram or showed the highest and lowest β phase stabilized alloys, respectively.

Fig. 3

OM images of (a) and (d) Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe, (b) and (e) Ti–4.5Cr–2.5Mn–1.1Al, (c) and (f) Ti–10.8Mo–2.3Sn–1.0Al alloys. (a), (b), (c) and (d), (e), (f) were obtained from as-cast and solution treated conditions, respectively. (1), (2) and (3) showing the paths and directions of point analysis measurement, corresponding to the ones in Fig. 5.

Fig. 4

X-ray diffraction profiles of (a) and (b) Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe, (c) and (d) Ti–4.5Cr–2.5Mn–1.1Al, (e) and (f) Ti–10.8Mo–2.3Sn–1.0Al alloys. (a), (c), (e) and (b), (d), (f) were obtained from as-cast and solution treated conditions, and (g) after tensile test for solution treated Ti–4.5Cr–2.5Mn–1.1Al alloy.

It was known that all three alloys showed slightly larger grain size at solution treated condition compared with their grain size at as-cast condition. The segment (1) and (2) or segment (2) and (3) in the images at both as-cast and solution treated conditions in Fig. 3, corresponded to the two adjacent grains, indicating the paths and directions of point analyses were from (1) to (3). The elemental point analyses were performed on these segments for the determination of concentration heterogeneity or solidification path. The distribution profiles of alloying elements in Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy at both as-cast and solution treated conditions were shown in Fig. 5(a) and (b), respectively. The higher and lower contents of solute alloying elements and Ti were observed with the position closer to grain boundary. For Ti–4.5Cr–2.5Mn–1.1Al and Ti–10.8Mo–2.3Sn–1.0Al alloys at as-cast condition as shown in Fig. 5(c) and (e), their solute-segregation conditions showed the same tendency with the as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy, and their solution treated specimens showed the homogeneous concentrations.

Fig. 5

The concentration profiles of alloying elements of (a) and (b) Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe, (c) and (d) Ti–4.5Cr–2.5Mn–1.1Al, (e) and (f) Ti–10.8Mo–2.3Sn–1.0Al alloys. (a), (c), (e) and (b), (d), (f) were obtained from as-cast and solution treated conditions, respectively. And the concentration profiles (a) corresponded to the Vickers hardness values in five positions. (1), (2) and (3) correspond to the ones in Fig. 3.

The degree of segregation of solute elements in a grain caused by dendritic growth correlated to their kinds and amount. In contrast, serious segregation in three β-Ti alloys could to be caused by solidification conditions.^37–39) It is considered in this CCLM process, that the solutes enrichment strongly depending on their diffusion coefficient in liquid was caused keeping the local equilibrium in liquid ahead of solid and liquid interface under different concentrations between solid and infinity in liquid representing by the effective equilibrium coefficient and a certain thickness of boundary layer depending on the solidification rate and stir of liquid in this solidification process. The concentration profiles of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy as shown in Fig. 5(a) corresponded to the initial, steady and final stages⁴⁰⁾ in solidification according to the solute enrichment model mentioned above, which resulted in heavy micro-segregation. The solidification proceeded from Ti enriched dendritic arms to Mn, Zr and Fe enriched parts corresponding to center to grain boundaries in grains, respectively, as shown in Fig. 5. In contrast, the micro Vickers hardness values were reflected the concentration profiles, as shown in Fig. 5(a), which meant the heavy segregation even in a grain of as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy.

For all solution treated specimens, the diffusion of alloying elements occurred due to the presence of elemental concentration gradients. There were constant values in all concentrations by solution heat treatment, and they were close with their chemical compositions, as shown in Figs. 1 and 5. For convenience, the segregation degree in a grain, K was denoted using the Ti content. In addition, Bo_{t max}, Bo_{t min}, ΔBo_t and Md_{t max}, Md_{t min}, ΔMd_t values of three β-Ti alloys at as-cast condition were also shown in Table 2. The segregation degree K was approximated by eq. (3) using C_{Ti max}, C_{Ti min} and C_{Ti total} showing concentrations of minimum, maximum and total Ti amounts, respectively.   

\begin{equation} K = (C_{\textit{Ti$\,$max}} - C_{\textit{Ti$\,$min}})/C_{\textit{Ti$\,$total}} \times 100\% \end{equation} (3)

Table 2 The segregation coefficient, K, Bo_{t max}, Bo_{t min}, ΔBo_t and Md_{t max}, Md_{t min}, ΔMd_t values of three β-Ti alloys at as-cast condition.

The K values, Bo_{t max}, Bo_{t min}, ΔBo_t and Md_{t max}, Md_{t min}, ΔMd_t values of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy with the most heavy segregation at as-cast condition were 8.3%, 2.795, 2.790, 0.005 and 2.330, 2.298, 0.032, respectively. The higher ΔBo_t and ΔMd_t may correlated to segregation degree, which results in the most heavy segregation degree of 4.8 times, compared with Ti–10.8Mo–2.3Sn–1.0Al with the lowest segregation. The relation between the segregation of alloying elements and positions of composition scatters in the Bo_t-Md_t diagram of as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy with the heavy segregation was also discussed in Fig. 6. According to the vectors of Ti–2 mass%M as shown in Fig. 1, it was known that the horizontal distribution of composition scatters in one grain was influenced by the changing in contents of Cr, Mn and Fe as shown in Fig. 6(a) and (c), corresponding to the diversification of Md_t values. In contrast, the vertical distribution of composition scatters was mainly influenced by the changing in the content of Zr as shown in Fig. 6(b), corresponding to the diversification of Bo_t values. Bo_t parameter describes the covalent bond strength between atoms, and correlates with several physical parameters of alloys such as the activation energy of diffusion in β-Ti alloys and melting points correlating to cohesion energy.¹¹⁾ Therefore, the higher concentration deviation was shown by the addition of Zr with higher Bo_t, which meant the slowly diffusion of Zr in liquid and solid parts in solidification process. The contents of gaseous impurities of oxygen and nitrogen were lower than those (oxygen and nitrogen: 0.069 and 0.006%, respectively) in the raw materials, because of highly vacuum level of 3 × 10⁻³ Pa as shown in Fig. 2. Moreover, it is found from the contents of impurities that the cleanly molten metals were created by utilization of CCLM without the reaction between the molten metal and water-cooled copper crucible, although the affinity of Ti with oxygen, carbon and nitrogen was strong.

Fig. 6

Solidification paths of (a) Ti, Cr and Mn, (b) Ti, Cr and Zr and (c) Ti, Cr and Fe for Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy at as-cast condition.

4.2 Tensile properties and fracture surface

The stress-strain curves of three β-Ti alloys at as-cast and solution treated conditions were shown in Fig. 7. Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy showing highest β phase stability with the lowest Md_t among three β-Ti alloys, had the slip behavior in the stress-strain curves showing a classical elasto-plastic behavior, and led to the high yield strength and negligible elongation. It had been reported that the tensile properties of β-Ti alloys depend significantly on the DT, such as slip and twin. Moreover, the content of the β phase stabilizing could greatly affect the formation of twin and slip.⁴¹⁾ There were high σ_0.2 more than 950 MPa, high σ_UTS more than 1000 MPa and ε_f more than 8%, regardless with heat treatment conditions. The fracture morphologies showing inter- and trans-granular or dimple fracture patterns, as shown in Fig. 8, that the same deformation behaviors were caused in both specimens, although there was their different segregation degree of solute elements, as seen in Fig. 5(a) and (b). A TEM image of the solution treated Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe specimen after fracture was shown in Fig. 9(a). Mono β phase was just observed even after tensile fracture. The dislocations corresponding to high flow stress levels, were already moved throughout β phase, which agreed with the predominant slip DT.

Fig. 7

Stress-strain curves of as-cast and solution treated three β-Ti alloys.

Fig. 8

Low and high magnified fractographies of tensile specimens for (a), (b) as-cast and (c), (d) solution treated Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe specimens, respectively.

Fig. 9

TEM images of (a) solution treated Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy at tensile fractured specimen, and solution treated Ti–4.5Cr–2.5Mn–1.1Al alloy at (b), (c) 1 and 5% plastic strained specimens, (d) SAED pattern along with $[\bar{1}11]$ of (c) and (e) tensile fractured specimen.

The as-cast and solution treated Ti–4.5Cr–2.5Mn–1.1Al alloys showed the σ_UTS value of 1033 and 804 MPa and σ_0.2 value of 803 and 394 MPa with ε_f value of 9 and 22%, respectively. There were large different flow stress levels between both specimens. The solution treated Ti–4.5Cr–2.5Mn–1.1Al showed typical four stages in the stress-strain curve. The first stage was the initial elastic region, the second stage was the transformation from a part of β phase to α′′ phase caused by the stress-induced transformation. The third stage corresponded to the elastic deformation of both β and α′′ phases. The last stage was the plastic deformation of β phase and α′′ phase. This stress-strain curve classified into four stages has been observed in Zr and Nb addition alloys.^15,42) The TEM images of the solution treated Ti–4.5Cr–2.5Mn–1.1Al specimen with two levels of 1 and 5% in plastic strain by cold rolling were shown in Fig. 9(b) and (c). There was just β phase in the 1% strained specimen. In contrast, some twin bands were observed obviously in 5% strained specimen. A selected-area electron diffraction pattern was taken from $[\bar{1}11]\beta $ zone axis and the reflections indicated the new part corresponding to the formation of twin bands were still β phase, as seen in Fig. 9(d), which agreed with previous literatures.^43–45) The tensile fractured samples showed some stress induced α′′ martensite phase in the twin band, as shown in Fig. 9(e), which agreed with to the predominant twin DT. Moreover, this agreed with the result of XRD showing additional new peaks of α′′ martensite phase, due to the formation of stress-induced α′′ martensite phase, as seen in Fig. 4(d) and (g).

The as-cast Ti–10.8Mo–2.3Sn–1.0Al specimen indicated the σ_UTS value of 600 MPa and σ_0.2 value of 520 MPa but large uniform ε_f of approximately 40%. In contrast, solution treated specimen indicated similar stress-strain behaviors to the as-cast one, showing the characteristic α′′ martensite deformation.^46,47) This agreed with to the predominant martensite DT, regardless of initial elastic parts. Experimental β-Ti alloys at as-cast and solution treated conditions showed the similar tensile behaviors. Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy with the slip DT showed the close stress-strain behaviors, regardless of the different segregation degrees of solute elements. In contrast, alloys with the twin and martensite DTs showed different behaviors in initial elastic region, because of the different precipitated α′′ martensite amount. Then, the predicted classification in β-Ti alloys by DT in the Bo_t-Md_t diagram, could be accuracy in three proposed β-Ti alloy systems.

4.3 Vickers hardness

The Vickers hardness of three β-Ti alloys at as-cast and solution treated conditions were shown in Fig. 10. The Vickers hardness values were 320 and 240 in as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al samples, respectively. Their solution treated specimens showed 3–5% lower Hv values, compared with as-cast ones. In contrast, solution treated Ti–4.5Cr–2.5Mn–1.1Al specimens showed 17% lower Hv value compared with as-cast one (330). They were similar to tensile behaviors, as shown in Fig. 7. All the three alloys showed higher hardness values at as-cast condition compared with their values at solution treated condition, which were attributed to the presence of segregation of solute elements. Moreover, it was clear that the linearity of four indentation edges was more accuracy of three solution treated specimens, compared with that of as-cast specimens, as shown in Fig. 11. The irregular indentation edges with expansion and poor symmetry of indentation might be caused by different forces from each direction due to different hardness. The difference in forces from each direction was attributed to the different solid solution strengthening due to the segregation behavior. This agreed with the correlation between segregation degree and corresponding hardness value for Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe with the highest segregation degree, as seen in Fig. 5(a), although there was its highest hardness value as mean one, as shown in Fig. 10.

Fig. 10

Vickers hardness of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe, Ti–4.5Cr–2.5Mn–1.1Al and Ti–10.8Mo–2.3Sn–1.0Al alloys for as-cast and solution treated specimens.

Fig. 11

Deformation bands around the Vickers indentations of (a) and (d) Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe, (b) and (e) Ti–4.5Cr–2.5Mn–1.1Al and (c) and (f) Ti–10.8Mo–2.3Sn–1.0Al alloys. (a), (b), (c) and (d), (e), (f) were obtained from as-cast and solution treated specimens, respectively.

The β-Ti alloys were deformed by either slip or twin mechanism, depending on the stability of the β phase.¹¹⁾ The wavy slip bands appear when the slip mechanism was dominant, whereas straight twin bands appear when the twin mechanism was operating. The deformation bands around Vickers indentation of three β-Ti alloys were also shown in Fig. 11. In Fig. 11(a) and (d), the observation results indicating the Vickers indentation of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy was wavy shape at both conditions, respectively. Moreover, the Vickers indentation around Ti–4.5Cr–2.5Mn–1.1Al alloy, as shown in Fig. 11(b) and (e), at both conditions exhibited straight twin bands which meant the twin mechanism was dominant. The Ti–10.8Mo–2.3Sn–1.0Al specimens with main twin bands were observed in Fig. 11(c) and (f). The Vickers indentation scale of Ti–10.8Mo–2.3Sn–1.0Al alloy was largest among three β-Ti alloys, which also could be seen as a feature for the martensite alloy. These Vickers indentation results of three β-Ti alloys were highly consistent with their DT in tensile tests.

4.4 As-cast application possibility of experimental alloys in the view of tension behaviors

The unit of fracture in tension as macro characterization was predominantly one grain even in the as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe having the highest segregation degree of solute elements as micro one. There was subsequent agreement in the stress-strain curves between its as-cast and solution treated samples. Other specimens showed a little difference in initial elastic regions of the curves, because of the different in the α′′ martensite amount. For instance, the mean hardness values were obtained in three alloys by the mixture rule which consisted of the corresponding hardness in some area showing different concentration profiles. The tensile and hardness properties as macro properties were defined by mean strength, elongation and hardness values of experimental alloys, although there was the lower linearity of indentation edges and poor symmetry of indentation. It was found that as-cast application possibility of both alloys of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al could not be refused in the view of tension behaviors. Promising alloys could be proposed accuracy in short periods using the Bo_t-Md_t diagram which showed high level in the prediction of classification in β-Ti alloys by DT.

5. Conclusions

1. (1)    Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy located in the slip DT showed mono β phase and the highest segregation degree of solute elements in experimental alloys. In contrast, Ti–4.5Cr–2.5Mn–1.1Al and Ti–10.8Mo–2.3Sn–1.0Al alloys located in the twin and martensite DTs showed the β phase and precipitated α′′ martensite, respectively, regardless of heat treatment conditions.
2. (2)    Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe alloy showing highest β phase stability with the lowest Md_t among three β-Ti alloys, had the slip behavior in the stress-strain curves showing a classical elasto-plastic behavior, and led to the high σ_0.2 more than 950 MPa, high σ_UTS more than 1000 MPa and ε_f more than 8%, regardless of the heat treatments.
3. (3)    The Vickers hardness values were 320 and 240 in as-cast Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al specimens, respectively. Their solution treated specimens showed 3–5% lower Hv values, compared with as-cast ones. Their hardness properties were similar to tensile behaviors.
4. (4)    It was found that as-cast application possibility of both alloys of Ti–5.5Cr–5.4Mn–5.1Zr–2.8Fe and Ti–10.8Mo–2.3Sn–1.0Al could not be refused in the view of tension behaviors. Promising alloys for as-cast applications could be proposed accuracy in short periods using the Bo_t-Md_t diagram which showed high level in the prediction of classification in alloys by DT.

Acknowledgements

This study was financially supported by JSPS KAKENHI C Grant Number 18K11712. Simultaneously, this study was financially supported by the Japan Foundry Engineering Society Found.

REFERENCES

• 1)   O.M.  Ivasishin,  P.E.  Markovsky,  Y.V.  Matviychuk,  S.L.  Semiatin,  C.H.  Ward and  S.  Fox: J. Alloy. Compd. 457 (2008) 296–309.
• 2)   J.  Ryhänen,  E.  Niemi,  W.  Serlo,  E.  Niemelä,  P.  Sandvik,  H.  Pernu and  T.  Salo: J. Biomed. Mater. Res. 35 (1997) 451–457.
• 3)   J.  Choi,  D.  Bogdanski,  M.  Koller,  S.A.  Esenwein,  D.  Muller,  G.  Muhr and  M.  Epple: Biomaterials 24 (2003) 3689–3696.
• 4)   K.  Matsugi,  H.  Kishimoto,  D.  Yamakawa,  Z.F.  Xu and  Y.B.  Choi: Mater. Trans. 56 (2015) 1747–1755.
• 5)   M.A.H.  Gepreel and  M.  Niinomi: J. Mech. Behav. Biomed. Mater. 20 (2013) 407–415.
• 6)   R.R.  Boyer and  R.D.  Briggs: J. Mater. Eng. Perform. 14 (2005) 681–685.
• 7)  I. Inagaki, T. Takechi, Y. Shirai and N. Ariyasu: Nippon Steel and Sumitomo Metal Technical Report, No. 106, (2014) 22–27.
• 8)  M. Peters, J. Hemptenmacher, J. Kumpfert and C. Leyens: Titanium and Titanium Alloys: Fundamentals and Applications, ed. by C. Leyens and M. Peters, (Wiley-Vch Verlag GmbH and Co. KGaA, 2003) pp. 1–36.
• 9)   E.O.  Ezugwu and  Z.M.  Wang: J. Mater. Process. Technol. 68 (1997) 262–274.
• 10)   F.H.  Froes,  H.  Friedrich,  J.  Kiese and  D.  Bergoint: JOM 56(2) (2004) 40–44.
• 11)  M. Morinaga, N. Yukawa, T. Maya, K. Sone and H. Adachi: Proc. 6th World Conf. on Ti, TMS, (1988) pp. 1601–1606.
• 12)   M.  Morinaga,  N.  Yukawa and  H.  Adachi: Tetsu-to-Hagané 72 (1986) 555–562.
• 13)   M.  Morinaga,  J.  Saito and  M.  Morishita: J. JILM 42 (1992) 614–621.
• 14)   P.  Laheurte,  F.  Prima,  A.  Eberhardt,  T.  Gloriant,  M.  Wary and  E.  Patoor: J. Mech. Behav. Biomed. Mater. 3 (2010) 565–573.
• 15)   X.J.  Jiang,  Y.K.  Zhou,  Z.H.  Feng,  C.Q.  Xia,  C.L.  Tan,  S.X.  Liang,  X.Y.  Zhang,  M.Z.  Ma and  R.P.  Liu: Mater. Sci. Eng. A 639 (2015) 407–411.
• 16)   C.P.  Liang and  H.R.  Gong: Int. J. Hydrogen Energ. 35 (2010) 3812–3816.
• 17)   Y.Y.  Zong,  S.H.  Huang,  Y.J.  Feng and  D.B.  Shan: J. Alloy. Compd. 541 (2012) 60–64.
• 18)   W.J.  Joost,  S.  Ankem and  M.M.  Kuklja: Acta Mater. 105 (2016) 44–51.
• 19)   J.R.S.  Martins,  J.R.O.  Araújo,  T.A.G.  Donato,  V.E.  Arana-Chavez,  M.A.R.  Buzalaf and  C.R.  Grandini: Materials 7 (2014) 232–243.
• 20)   R.  Sarkar,  P.  Ghosal,  T.K.  Nandy and  K.K.  Ray: Mater. Sci. Eng. A 656 (2016) 223–233.
• 21)   A.  Gagnoud,  J.  Etay and  M.  Garnier: Trans. ISIJ 28 (1988) 36–40.
• 22)   A.  Morita,  H.  Fukui,  H.  Tadano,  S.  Hayashi,  J.  Hasegawa and  M.  Niinomi: Mater. Sci. Eng. A 280 (2000) 208–213.
• 23)   K.  Matsugi,  H.  Mamiya,  Y.B.  Choi,  G.  Sasaki,  O.  Yanagisawa and  H.  Kuramoto: Int. J. Cast. Met. Res. 21 (2008) 156–161.
• 24)   C.  Brozek,  F.  Sun,  P.  Vermaut,  Y.  Millet,  A.  Lenain,  D.  Embury,  P.J.  Jacques and  F.  Prima: Scr. Mater. 114 (2016) 60–64.
• 25)  M. Morinaga, M. Kato, T. Kamimura, M. Fukumoto, I. Harada and K. Kubo: Proc. 7th World Conf. on Ti, TMS, (1992) pp. 217–224.
• 26)   K.  Ikeda: J. JILM 55 (2005) 97–102.
• 27)   I.V.  Okulov,  H.  Wendrock,  A.S.  Volegov,  H.  Attar,  U.  Kühn,  W.  Skrotzki and  J.  Eckert: Mater. Sci. Eng. A 628 (2015) 297–302.
• 28)   B.  Cantor,  I.T.H.  Chang,  P.  Knight and  A.J.B.  Vincent: Mater. Sci. Eng. A 375–377 (2004) 213–218.
• 29)   V.B.  Deev,  V.A.  Degtyar,  A.I.  Kutsenko,  I.F.  Selyanin and  A.P.  Voitkov: Steel Transl. 37 (2007) 991–994.
• 30)   A.K.  Dahle,  Y.C.  Lee,  M.D.  Nave,  P.L.  Schaffer and  D.H.  StJohn: J. Light Met. 1 (2001) 61–72.
• 31)   M.  Morinaga: Mater. Trans. 57 (2016) 213–226.
• 32)   R.J.  Deeth: J. Chem. Soc., Dalton Trans. 0 (1990) 355–363.
• 33)   K.  Matsugi,  T.  Endo,  Y.B.  Choi and  G.  Sasaki: Mater. Trans. 51 (2010) 740–748.
• 34)   M.  Abdel-Hady,  H.  Fuwa,  K.  Hinoshita,  H.  Kimura,  Y.  Shinzato and  M.  Morinaga: Scr. Mater. 57 (2007) 1000–1003.
• 35)   D.  Kuroda,  M.  Niinomi,  M.  Morinaga,  Y.  Kato and  T.  Yashiro: Mater. Sci. Eng. A 243 (1998) 244–249.
• 36)   Y.  Murayama,  S.  Sasaki,  H.  Kimura and  A.  Chiba: Mater. Sci. Forum 638–642 (2010) 635–640.
• 37)   A.K.  Sample and  A.  Hellawell: Metall. Trans. A 15 (1984) 2163–2173.
• 38)   D.  Ruvalcaba,  R.H.  Mathiesen,  D.G.  Eskin,  L.  Arnberg and  L.  Katgerman: Acta Mater. 55 (2007) 4287–4292.
• 39)   H.  Kato and  J.R.  Cahoon: Metall. Trans. A 16 (1985) 579–587.
• 40)  W. Kurz and D.J. Fisher: Fundamentals of Solidification, (Trans Tech Publications, Zurich, 1998) pp. 117–132.
• 41)   X.H.  Min,  K.  Tsuzaki,  S.  Emura,  T.  Nishimura and  K.  Tsuchiya: Mater. Trans. 52 (2011) 1611–1616.
• 42)   A.  Paradkar,  S.V.  Kamat,  A.K.  Gogia and  B.P.  Kashyap: Mater. Sci. Eng. A 487 (2008) 14–19.
• 43)   J.H.  Gao,  Y.H.  Huang,  D.K.  Guan,  A.J.  Knowles,  L.  Ma,  D.  Dye and  W.M.  Rainforth: Acta Mater. 152 (2018) 301–314.
• 44)   L.  Ren,  W.L.  Xiao,  C.L.  Ma,  R.X.  Zheng and  L.  Zhou: Scr. Mater. 156 (2018) 47–50.
• 45)   F.  Sun,  J.Y.  Zhang,  M.  Marteleur,  T.  Gloriant,  P.  Vermaut,  D.  Laille,  P.  Castany,  C.  Curfs,  P.J.  Jacques and  F.  Prima: Acta Mater. 61 (2013) 6406–6417.
• 46)   L.L.  Chang,  Y.D.  Wang and  Y.  Ren: Mater. Sci. Eng. A 651 (2016) 442–448.
• 47)   H.Y.  Kim,  Y.  Ikehara,  J.I.  Kim,  H.  Hosoda and  S.  Miyazaki: Acta Mater. 54 (2006) 2419–2429.