Elsevier

Acta Materialia

Volume 59, Issue 18, October 2011, Pages 6975-6988
Acta Materialia

Low Young’s modulus in Ti–Nb–Ta–Zr–O alloys: Cold working and oxygen effects

https://doi.org/10.1016/j.actamat.2011.07.050Get rights and content

Abstract

The origin of the low Young’s modulus of cold worked Ti–36Nb–2Ta–3Zr–xO mass% polycrystals with a body-centered cubic (β-phase) structure, referred to as gum metal, was investigated with a focus on the roles of oxygen concentration, the electron–atom (e/a) ratio, and the cold working process. Analysis of the temperature dependence of the microstructures and elastic properties of single crystals at x = 0.09, 0.36, 0.51% O using transmission electron microscopy and an electromagnetic acoustic resonance method, respectively, revealed that the shear moduli c′ and c44 of the 0.36 and 0.51% O alloys softened upon cooling near room temperature (RT) and exhibited low values at RT. This was because suppression of the α″ martensitic transformation by oxygen addition led to retention of the low stability single β-phase state at RT. The Hill approximation indicated that the low c′ and c44 values caused by softening gave rise to the low Young’s modulus, which is common to some Ti–Nb-based alloys with an e/a ratio of ∼4.24. Analysis of the microstructures and elastic properties of solution-treated and cold worked x = 0.06, 0.30, 0.47% O alloy polycrystals at RT revealed that the Young’s modulus increased upon 90% cold working due to formation of the α″ martensite phase (0.09% O) and ω phase (0.09, 0.30, and 0.47% O) with a high elastic modulus in the β-phase matrix. However, increasing the oxygen concentration suppresses the increase in Young’s modulus because oxygen addition decreases the amount of α″ and ω phases formed while retaining the low stability β phase. Therefore, cold working combined with oxygen addition produces a low Young’s modulus compatible with high strength.

Introduction

Recently a class of Ti alloys termed gum metal has attracted considerable attention because of their high strength (∼1.1 GPa), yield strain, and ductility, in addition to their low Young’s modulus [1], [2], [3], [4]; it should be noted that a low Young’s modulus is required for application as biomedical implants, to prevent bone degradation and absorption caused by the difference in Young’s modulus of the implant and natural human bone [5]. This unique combination of superior properties is obtained by cold working (∼90%) body-centered cubic (β-phase) Ti alloys with a certain amount of oxygen and an electron–atom ratio (e/a) of ∼4.24. One typical Gum metal alloy composition is Ti–36Nb–2Ta–3Zr–0.3O in mass%. Previous studies have suggested that a low Young’s modulus is closely related to a unique combination of superior mechanical properties [1], [6]. However, the origin of the low Young’s modulus is still unknown, because the effects of important processing parameters (e/a ratio, oxygen addition, and cold working) on the elastic properties have not been fully elucidated.

The elastic properties of single crystals of gum metal have been a particular focal point [7], [8], [9], [10], [11], because it has been claimed that the shear modulus c′ for {1 1 0}11¯0 shear approaches zero near an e/a value of 4.24, and it is this anomalously low c′ that gives rise to the low Young’s modulus [1], [6]. These claims are based on the relationship between c′ and e/a values estimated by a first principles calculation based on density functional theory (DFT) [7]. On the other hand, Talling et al. employed cold worked “polycrystalline” gum metal specimens and extracted the c′ value of a “single crystal” in the cold worked state, on the basis of diffraction strain measurements using in situ synchrotron X-ray diffraction [9], [10]. Their study suggested that the c′ value in the cold worked state is not anomalously low and is almost the same as the c′ values of conventional binary Ti–V alloys with similar e/a ratios. Takesue et al. successfully grew single crystals of gum metal [11] and measured the Young’s modulus of the solution-treated single crystals. However, their measurements were limited to specific crystallographic directions. Thus the effect of the e/a ratio on c′ in a single crystal has not been fully determined. Consequently, to clarify whether or not the c′ value of gum metal is anomalously low it is crucial to obtain direct measurements of all the independent elastic stiffness variables of a single crystal.

The concentration of added oxygen and the cold working conditions, as well as the e/a ratio, are important in the processing of gum metal, and however, their effects on the elastic properties have not yet been clarified. Although a strong 〈1 1 0〉 fiber texture is found in cold worked (cold swaged) polycrystalline gum metal [3], [12] the Young’s modulus has only been measured in the fiber axis (swaging direction) [1], [8]. In addition to the fiber texture, a nanosized ω (hexagonal) phase is formed in the β phase matrix as a result of cold working [13], [14]. It was suggested that the formation of a nanosized ω phase contributes to the high strength of gum metal alloys. However, the effect of this ω phase formation on the elastic properties of these alloys has also not yet been investigated. It is also important to clarify the effect of oxygen concentration on the elastic properties of the alloys because addition of oxygen suppresses formation of the ω and α″ martensitic phases [15], [16], [17], [18], [19], which strongly affect the elastic properties of β phase Ti alloys, in addition to enhancing the mechanical strength [20].

In the present study the origin of the low Young’s modulus of cold worked Ti–Nb–Ta–Zr–O alloys (gum metal) was investigated, with a focus on the effects of three important processing features on the elastic properties: e/a ratio, oxygen addition, and cold working. In order to fully elucidate the effects of the e/a ratio and oxygen concentration on the elastic properties single crystals of Ti–36Nb–2Ta–3Zr–(0.09, 0.36, 0.51)O mass% alloys were grown. The temperature and oxygen concentration dependence of the microstructures and the elastic properties of the solution-treated single crystals were investigated using transmission electron microscopy (TEM) and an electromagnetic acoustic resonance (EMAR) method [21], respectively. The measured elastic properties were systematically compared with those of binary Ti-based alloys on the basis of the e/a ratio, to reveal the unique elastic characteristics of Ti–Nb–Ta–Zr–O alloys. In addition to the solution-treated single crystals, cold worked Ti–36Nb–2Ta–3Zr–(0.06, 0.30, 0.47)O mass% alloy polycrystals were prepared. The texture formed by the cold working process was analyzed using X-ray pole figures, and the complete set of anisotropic elastic constants was measured by resonant ultrasound spectroscopy. The measured elastic constants were compared with those of solution-treated polycrystals and, thereby, the effects of cold working and oxygen concentration on the elastic properties were elucidated. Finally, the origin of the low Young’s modulus in cold worked gum metal was discussed, taking into account the effects of e/a ratio, oxygen concentration, and cold working on the elastic properties.

Section snippets

Experimental procedure

Master alloy bars with nominal compositions Ti–36Nb–2Ta–3Zr–(0.3, 0.5)O mass% were prepared by a powder metallurgical method consisting of cold isostatic pressing, sintering, and hot forging [13]. The master alloy bars were solution-treated at either 1173 K (0.3% O) or 1323 K (0.5% O) for 1.8 ks, under vacuum, and quenched in water. After the solution treatment the cold worked bars were prepared by cold working with a rotary swaging machine to obtain an ∼90% reduction in area. In addition, single

Microstructures of solution-treated single crystals by TEM observation

Fig. 1a shows a bright field image of the microstructure of a 0.09% O alloy single crystal at room temperature, where the beam direction is parallel to [1 1 0]β (i.e. the [1 1 0] direction in the β phase). The bright field image indicates that band-like products are formed in the β phase matrix. The corresponding electron diffraction pattern and key diagram, shown in Fig. 1b and c, respectively, indicate that in addition to the fundamental spots derived from the β phase, additional spots derived

β-Phase stability of solution-treated Ti–Nb–Ta–Zr–O alloys

Zener [49] insisted that the shear modulus c′ is a direct indicator of β phase stability, where a low c′ corresponds to low β phase stability. Based on this assertion, a comparison of the c′ values of Ti–Nb–Ta–Zr–O and binary Ti-based alloys indicates that the β phase stability of Ti–Nb–Ta–Zr–O is remarkably lower than that of other binary Ti-based alloys. Thus it is reasonable that the low β phase stability promotes ω phase formation during cold working of these alloys. In the case of the 0.36

Conclusions

The elastic properties of Ti–36Nb–2Ta–3Zr–(0.09, 0.36, 0.51)O mass% alloy single crystals with a bcc (β phase) structure and those of solution-treated and cold worked Ti–36Nb–2Ta–3Zr–(0.06, 0.30, 0.47)O mass% alloy polycrystals were investigated to clarify the origin of the low Young’s modulus of the cold worked alloy polycrystals (gum metal). The conclusions are summarized as follows.

  • i.

    The observed increase upon cooling of the shear moduli c′ and c44 of the 0.09% O alloy single crystal

Acknowledgements

The authors thank Prof. H. Yasuda and Mr. E. Taguchi of the Research Center for Ultra-high Voltage Electron Microscopy and Dr. Hagihara of the Graduate School of Engineering, Osaka University, for the TEM observations. This study was supported by Priority Assistance for the Formation of Worldwide Renowned Centers of Research – The Global COE Program (Project: Center of Excellence for Advanced Structural and Functional Materials Design) and a Grant-in-Aid for Scientific Research and Development

References (63)

  • E.A. Withey et al.

    Acta Mater

    (2010)
  • J.W. Morris et al.

    Acta Mater

    (2010)
  • M.Y. Gutkin et al.

    Acta Mater

    (2006)
  • M. Geetha et al.

    Prog Mater Sci

    (2009)
  • S. Kuramoto et al.

    Mater Sci Eng A

    (2006)
  • R.J. Talling et al.

    Scripta Mater

    (2008)
  • R.J. Talling et al.

    Scripta Mater

    (2009)
  • N. Takesue et al.

    J Cryst Growth

    (2009)
  • T. Yano et al.

    Acta Mater

    (2009)
  • T. Yano et al.

    Scripta Mater

    (2010)
  • J.I. Qazi et al.

    Mater Sci Eng C

    (2005)
  • N.E. Paton et al.

    Scripta Metall

    (1973)
  • M. Tane et al.

    Acta Mater

    (2008)
  • M. Tane et al.

    Acta Mater

    (2010)
  • S.K. Sikka et al.

    Prog Mater Sci

    (1982)
  • I.B. Ramsteiner et al.

    Acta Mater

    (2008)
  • Y. Al-Zain et al.

    Acta Mater

    (2011)
  • H.Y. Kim et al.

    Mater Sci Eng A

    (2006)
  • R.J. Talling et al.

    Acta Mater

    (2009)
  • E.S. Fisher et al.

    Acta Metall

    (1970)
  • L.A. Ahlberg et al.

    Scripta Metall

    (1978)
  • T. Ichitsubo et al.

    Acta Mater

    (2002)
  • K. Otsuka et al.

    Prog Mater Sci

    (2005)
  • A. Planes et al.

    Solid State Phys

    (2001)
  • E. Walker

    Solid State Commun

    (1978)
  • M. Nakai et al.

    Mater Lett

    (2011)
  • S.Q. Zhang et al.

    Acta Mater

    (2011)
  • H.M. Ledbetter et al.

    Mater Sci Eng A

    (1988)
  • T. Saito et al.

    Science

    (2003)
  • T. Li et al.

    Phys Rev Lett

    (2007)
  • H. Ikehata et al.

    Phys Rev B

    (2004)
  • Cited by (123)

    View all citing articles on Scopus
    View full text