Strengthening mechanisms in Ti-Nb-Zr-Ta and Ti-Mo-Zr-Fe orthopaedic alloys

Rajarshi Banerjee ^, , Soumya Nag , John Stechschulte and Hamish L. Fraser

Department of Materials Science and Engineering, The Ohio State University, 477 Watts Hall, 2041 College Road, Columbus, OH 43210, USA

Received 22 September 2003;  accepted 11 October 2003.  Available online 26 November 2003.
Biomaterials
Volume 25, Issue 17 , August 2004, Pages 3413-3419

1. Abstract

The microstructural evolution and attendant strengthening mechanisms in two novel orthopaedic alloy systems, Ti-Nb-Zr-Ta and Ti-Mo-Zr-Fe, have been compared and contrasted in this paper. Specifically, the alloy compositions considered are Ti-34Nb-9Zr-8Ta and Ti-13Mo-7Zr-3Fe. In the homogenized condition, both alloys exhibited a microstructure consisting primarily of a
matrix with grain boundary
precipitates and a low-volume fraction of intra-granular
precipitates. On ageing the homogenized alloys at 600°C for 4 hr, both alloys exhibited the precipitation of refined scale secondary
precipitates homogeneously in the
matrix. However, while the hardness of the Ti-Mo-Zr-Fe alloy marginally increased, that of the Ti-Nb-Zr-Ta alloy decreased substantially as a result of the ageing treatment. In order to understand this difference in the mechanical properties after ageing, TEM studies have been carried out on both alloys prior to and post the ageing treatment. The results indicate the existence of a metastable B2 ordering in the Ti-Nb-Zr-Ta alloy in the homogenized condition which is destroyed by the ageing treatment, consequently leading to a decrease in the hardness.

Author Keywords: Author Keywords: Titanium alloy; Microstructure; TEM

2. Article Outline

1. Introduction

2. Experimental procedure

3. Results

3.1. TNZT alloy

3.2. TMZF alloy

4. Discussion

5. Summary and conclusions

References

3. 1. Introduction

The ideal biomaterial for implant applications, especially for joint replacements, is expected to exhibit excellent properties such as no adverse tissue reactions, excellent corrosion resistance in the body fluid medium, high mechanical strength and fatigue resistance, low modulus, low-density, and good wear resistance [1, 2, 3 and 4]. In the past decade, there have been some efforts directed towards development of Ti-base biomaterials specifically geared towards implant applications. The first generation of these orthopaedic alloys included Ti-6Al-7Nb [5] and Ti-5Al-2.5Fe [6]. These two alloys exhibited mechanical properties similar to Ti-6Al-4 V and were primarily developed in response to concerns of potential cytotoxicity and adverse tissue reactions caused by V [7 and 8]. Further studies have shown the release of both V and Al ions from the alloy might cause long-term health problems, such as peripheral neuropathy, osteomalacia, and Alzheimer diseases [9 and 10]. Subsequent developments in orthopaedic Ti-base alloys have been largely motivated by the requirement for low modulus materials. Finite-element simulations suggest that lower modulus material used for joint replacements may better simulate the natural femur in distributing stress to the adjacent bone tissue [11 and 12]. Alloys such as Ti-6Al-4 V, which consist of a relatively large volume fraction of the hcp
phase mixed with a smaller volume fraction of the bcc
phase, exhibit a modulus ~110 GPa which is substantially higher than that of bone tissue (~10-40 GPa) [1]. Such a large modulus mismatch causes insufficient loading of the bone adjacent to the implant. This is often referred to as the stress-shielding phenomenon and can lead to potential bone resorption and eventual failure of the implant [13]. The motivation to develop lower modulus alloys has lead to an increased focus on
-Ti alloys which retain a single-phase
microstructure on rapidly cooling from high temperatures. Metastable
-Ti alloys developed for this purpose include, Ti-12Mo-6Zr-2Fe, also referred to as `TMZF' [14], Ti-15Mo-5Zr-3Al [15], Ti-15Mo-3Nb-3O, also referred to as TIMETAL 21SRx [16], and Ti-13Nb-13Zr [17]. The lowest elastic modulus that has been achieved to-date is for the `TNZT' alloys based on the Ti-Nb-Zr-Ta system and a promising composition in this system is Ti-35Nb-7Zr-5Ta [18].

In the present paper, the microstructural evolution in two recently developed low modulus
-Ti alloys, based on the Ti-Nb-Zr-Ta and Ti-Mo-Zr-Fe systems, has been investigated in detail using a combination of scanning and transmission electron microscopes. The influence of the microstructural evolution on the hardness and modulus of these alloys in the homogenized and homogenized+aged conditions have also been studied in detail. For convenience, the Ti-Nb-Zr-Ta alloy will be subsequently referred to as TNZT and the Ti-Mo-Zr-Fe alloy as TMZF.

4. 2. Experimental procedure

The TNZT alloy was prepared by melting 5.3 gm CP titanium, 3.4 gm niobium (99.95% pure), 0.6 gm zirconium (99.5% pure), and 0.8 gm tantalum (99.95% pure) in a vacuum arc melter (manufactured by TEK Specialities INC, Winchester, MA). Similarly for melting TMZF, 8 gm CP Ti, 1.2 gm molybdenum (99.8% pure), 0.6 gm zirconium (99.5% pure) and 0.2 gm iron (99.97% pure) were used. The melting chamber was first evacuated and then back-filled with argon gas (pressure ~13 in Hg). Buttons, approximately 10 gms each, were melted in a copper hearth with a tungsten electrode. The buttons were re-melted four times in order to ensure chemical homogeneity. The arc-melted buttons were homogenized by heat-treating at 1100°C for seven days in a LINDBERG box furnace and then furnace cooled. Subsequently, the alloys were aged at 600°C for 4 hr and air-cooled. The homogenized and homogenized plus aged samples were mounted and mechanically polished using standard metallographic procedures.

Scanning electron microscopy (SEM) of these samples was performed using an FEI Sirion system equipped with a field emission gun (FEG) emitter and an EDS detector. Transmission electron microscopy (TEM) was performed using a FEI/Philips CM200 TEM system operating at 200 kV accelerating voltage. Electron-transparent TEM specimens were prepared from site specific locations using the focussed ion beam system (FEI Dual Beam 235 FIB). Microhardness tests were performed on these samples using a Vickers Micro Hardness Tester. The samples were tested using a load of 500 gf applied for 15 sec. In addition to microhardness testing, in order to evaluate the modulus of these alloys, they were tested using nanoindentation in a Nanoindenter XP system from MTS Instruments. A Berkovich diamond indenter was used and measurements carried out for 2 
m deep indents. The modulus and hardness were evaluated from the unloading portion of the load-displacement data using standardized routines available in the MTS Testworks software.

5. 3. Results

The average compositions of the TNZT alloy as measured using EDS in the SEM was Ti-34Nb-9Zr-8Ta and that of the TMZF alloy was Ti-13Mo-7Zr-3Fe.

3.1. TNZT alloy

The overall microstructure of the TNZT alloy in the homogenized condition is shown in the SEM backscatter image in Fig. 1(a). The microstructure consists of relatively large grains of the
phase with grain boundary
precipitates. A higher magnification image of the grain boundary
precipitates is shown in Fig. 1(b). In addition, some amount of intra-granular
precipitation is also visible within the
grains in Fig. 1(a). These intra-granular primary
precipitates often exhibit rather interesting morphologies as shown in Fig. 1(c). The average microhardness of the TNZT alloy in the homogenized condition is 291 VHN. The modulus as measured using nanoindentation was 100 GPa. The overall microstructure of the homogenized TNZT alloy after ageing at 600°C for 4 hr is shown in Fig. 2(a). An increased volume fraction of the intra-granular primary
precipitates is visible in this figure as compared to Fig. 1(a). However, in addition to these coarser intra-granular
precipitates, a finer scale secondary
precipitation occurs after the ageing treatment as shown in the higher magnification image in Fig. 2(b). These finer scale
precipitates are homogeneously distributed throughout the
matrix. However, it should be noted that there are precipitate-free zones surrounding the grain boundary
precipitates and also the coarser intra-granular primary
precipitates, visible in Fig. 2(b). These precipitate-free zones do not contain the fine scale secondary
. The average microhardness of the homogenized TNZT alloy after ageing is 258 VHN while the modulus measured using nanoindentation is 89 GPa. The reduction in the microhardness and consequently the strength of the TNZT alloy after ageing is rather surprising since the homogeneous precipitation of the fine scale
after ageing is expected to increase the strength of the alloy. Therefore, further detailed investigations of the microstructural evolution in this alloy were carried out using transmission electron microscopy (TEM).

(36K)

Fig. 1. SEM backscatter images of the homogenized TNZT sample: (a) overall microstructure showing large grains of
-Ti; (b) higher magnification image showing the grain boundary
precipitates; (c) intragranular primary
precipitates.

(53K)

Fig. 2. SEM backscatter images of the homogenized and aged TNZT sample: (a) overall microstructure showing large grains of
-Ti; (b) higher magnification image showing the intragranular primary
precipitates and fine scale secondary
forming as a result of the ageing treatment.

A bright-field TEM micrograph of the TNZT alloy in the homogenized condition, exhibiting the
matrix with a few small
precipitates, is shown in Fig. 3(a). This TEM sample was prepared from a region within a
grain that contained neither any coarse intra-granular primary
precipitates nor any grain boundary
precipitates. Selected area diffraction (SAD) patterns from the
matrix are shown in Figs. 3(b) and (c), which can be consistently indexed as the [0 0 1] and [0 1 1] zone axes respectively. In the [0 0 1]
pattern, in addition to the primary reflections arising from the matrix, additional reflections are visible at the 1/2 {1 1 0} and 1/2 {2 0 0} positions. The reflections at 1/2 {1 1 0} can be attributed to the precipitation of very fine scale tertiary
within the
matrix [19]. The scale of this
is so refined that it is difficult to image them as discrete precipitates even in the TEM and often appears as a strain contrast within the matrix. The reflections at the 1/2 {2 0 0} positions cannot be attributed to the secondary precipitation of either
or any other metastable phase (such as
which does not give rise to any additional reflections in the [0 0 1]
pattern). These reflections arise solely due to B2 type chemical ordering in the
matrix. These are the {1 0 0} superlattice reflections arising from B2 ordering. These spots are usually very sharp as evidenced both in the SAD pattern and the intensity line profile along g={2 0 0} vector shown in Fig. 3(b). In the [0 1 1]
pattern shown in Fig. 3(c), in addition to the primary
reflections, spots are visible at the 1/3 {1 1 2} and 1/2 {1 1 2} positions. While the spots at 1/2 {1 1 2} arise from the highly refined scale secondary
precipitation, the spots at 1/3 {1 1 2} arise from the precipitation of
in the
matrix. These additional reflections are clearly visible in the intensity line profile along the g={2 1−1} vector shown in Fig. 3(c). In addition, sharp superlattice {1 0 0} spots, attributable to B2 ordering, are also visible in this diffraction pattern.

(43K)

Fig. 3. (a) Bright-field TEM image showing the
-matrix in the homogenized TNZT alloy; (b) [0 0 1]
SAD pattern and intensity profile along g=200 reciprocal lattice vector; (c) [0 1 1]
SAD pattern and intensity profile along g=2 1−1 reciprocal lattice vector.

After ageing at 600°C for 4 hr, the microstructure within the
matrix is shown in the bright-field micrographs, Figs. 4(a) and (b). Fine scale secondary
precipitates which formed as a result of the ageing treatment, are visible in these micrographs. These precipitates are of the same type as observed earlier in the SEM studies on the aged TNZT alloy (refer Fig. 2(b)). SAD patterns of the [0 0 1]
and [0 1 1]
zone axes are shown in Figs. 4(c) and (d). These diffraction patterns have been recorded from regions of the matrix which did not include any fine scale secondary
precipitates. In addition to the primary
reflections, additional spots are visible in Fig. 4(c) at the 1/2 {1 1 0} positions corresponding to the very fine scale tertiary
discussed previously. Interestingly, no additional reflections can be observed at the 1/2 {2 0 0} positions in Fig. 4(c) as confirmed by the intensity line profile along g={2 0 0} shown in the same figure. This indicates that the matrix is no longer chemically ordered with B2-type ordering. The [0 1 1]
SAD pattern shown in Fig. 4(d), exhibits additional reflections at the 1/3 {1 1 2} and 1/2 {1 1 2} positions corresponding to the precipitation of
and the very fine scale
, respectively. The intensity line profile along g={2 1−1} vector, also shown in Fig. 4(d), confirms the same.

(42K)

Fig. 4. (a) and (b) Bright-field TEM images showing the fine-scale secondary
within the
-matrix in the homogenized and aged TNZT alloy; (b) [0 0 1]
SAD pattern and intensity profile along g=200 reciprocal lattice vector; (c) [0 1 1]
SAD pattern and intensity profile along g=2 1−1 reciprocal lattice vector.

3.2. TMZF alloy

In the case of the TMZF alloy, the overall microstructure of the alloy in the homogenized condition is shown in the SEM backscatter image (Fig. 5(a)). Similar to the case of the homogenized TNZT alloy, the microstructure consists primarily of large
grains with grain boundary
precipitates and some intra-granular
precipitation at a relatively coarser scale. The grain boundary
precipitates and the intra-granular
are visible in the higher magnification SEM image shown in Fig. 5(b). The average microhardness of the homogenized TMZF alloy is 345 VHN and the modulus measured by nanoindentation is 89 GPa. Similar to the case of the TNZT alloy, the microstructure of the TMZF alloy after ageing at 600°C for 4 hr exhibited the homogeneous precipitation of a fine scale secondary
. The overall homogenized plus aged microstructure of TMZF is shown in Fig. 6(a). A
grain boundary decorated with
precipitates, the intragranular
precipitates, and the fine scale secondary
precipitates are shown in Fig. 6(b). It should be noted that precipitate-free zones are present on either side of the grain boundary for both the intragranular
as well as the fine scale secondary
. The extent of the precipitate-free zone is greater for the relatively coarser intra-granular
as compared with that for the fine scale secondary
, as seen in Fig. 6(b). Additionally, the intra-granular
precipitates also exhibit precipitate-free zones which are devoid of any fine scale secondary
precipitation, as shown in Figs. 6(b) and (c). Microhardness of the homogenized plus aged TMZF is 366 VHN while the modulus measured using nanoindentation is 118 GPa.

(49K)

Fig. 5. SEM backscatter images of the homogenized TMZF sample: (a) overall microstructure showing grains of
-Ti; (b) higher magnification image showing the grain boundary
precipitates and intragranular primary
precipitates.

(83K)

Fig. 6. SEM backscatter images of the homogenized and aged TMZF sample: (a) overall microstructure showing grains of
-Ti; (b) higher magnification image showing the grain boundary
precipitates; (c) higher magnification image showing the intragranular primary
precipitates and fine scale secondary
forming as a result of the ageing treatment. Note the precipitate-free zone next to the grain boundary
and the intragranular primary
precipitates.

Details of the microstructural evolution in the TMZF alloy, in both homogenized as well as aged conditions, have been studied using TEM. Thus, a SAD pattern from the
matrix (not including any intra-granular
) of the homogenized TMZF alloy, which can be consistently indexed as the [0 1 1]
zone axis, is shown in Fig. 7(a). In addition to the primary reflections arising from the
matrix, spots at 1/3 {1 1 2} locations are visible which indicate the presence of
precipitates in this alloy. It should be noted that the additional rings in this diffraction pattern arise from the carbon film supporting the TEM specimen. A [0 1 1]
SAD pattern from the same alloy after ageing is shown in Fig. 7(b). In this case too additional reflections at the 1/3 {1 1 2} positions indicates the presence of
precipitates. Intensity profiles along the g={2 1−1} vector on each of these diffraction patterns are shown below the respective patterns and substantiate the presence of the peak arising from the
phase. The precipitation of the fine scale secondary
in the
matrix as a result of the ageing heat treatment is shown in the bright field TEM micrograph (Fig. 7(c)). The intensity of
reflections after ageing is similar to that observed in the homogenized condition (cf. Figs. 7(a) and (b)).

(28K)

Fig. 7. (a) [0 1 1]
SAD pattern and intensity profile along g=2 1−1 reciprocal lattice vector in the homogenized TMZF alloy. Note that the intensity profile shows only one of the
reflections; (b) [0 1 1]
SAD pattern and intensity profile along g=2 1−1 reciprocal lattice vector in the homogenized and aged TMZF alloy; (c) Bright-field TEM image showing the fine-scale secondary
within the
-matrix in the homogenized and aged TMZF alloy.

6. 4. Discussion

The two
-Ti alloys investigated in the present study exhibit an interesting contrast in terms of their strengthening mechanisms and associated phase transformations. In case of the TNZT alloy, the homogenized state exhibits a metastable B2 ordering in the matrix which leads to higher hardness and modulus values. On ageing the matrix disorders to
and fine scale secondary
precipitates in the matrix. These observations suggest that the precipitation of the secondary
might be responsible for the B2 to
transformation in the matrix due to the enrichment of the matrix with
stabilizing elements (Nb and Ta in this case) on the precipitation of
. It is, however, interesting to note that though the precipitation of fine scale secondary
is expected to aid in strengthening the alloy, the B2 to
transformation has a more dominant effect on the hardness of this alloy so that it softens as a result of the ageing treatment. A similar trend is observed in the experimentally measured modulus values with the homogenized condition exhibiting a higher modulus as compared with the aged condition. Precipitation of the
phase is observed in both the homogenized as well as the aged conditions, with the relative intensities of 1/3 {1 1 2} reflections being either similar in both conditions or exhibiting a marginal increase in case of the aged condition. Thus, the strengthening of the TNZT alloy from the
precipitates appears to be relatively similar for both the homogenized and aged conditions. It should be noted that similar to the observations presented here for the Ti-34Nb-9Zr-8Ta alloy, Tang, Ahmed, and Rack [19] have reported a higher hardness value in the homogenized condition as compared with homogenized plus heat-treated condition for the Ti-35Nb-7Zr-5Ta alloy. However, the reasons for the higher hardness in the homogenized condition have not been discussed by Tang et al. in their paper [19].

While the B2 to
transformation appears to be playing the dominant role amongst the strengthening mechanisms operating in the TNZT alloy, a different mechanism operates in case of the TMZF alloy. Thus, the TMZF alloy exhibits a higher hardness in the aged condition as compared with the homogenized condition. No B2 ordering is observed in the matrix of this alloy in either the homogenized or aged conditions. The precipitation of
is observed in both conditions, with the relative intensity of the 1/3 {1 1 2} reflections being the same or marginally greater in case of the aged condition. Therefore, it appears that in this case too, the
precipitates play a similar role in strengthening the alloy in both homogenized as well as aged conditions. The precipitation of the fine scale secondary
is the major microstructural variation accompanying the ageing heat treatment post homogenization. Therefore, the marginal increase in hardness of the TMZF alloy due to ageing can be attributed to the precipitation of fine scale secondary
. A similar trend is observed in the case of the modulus wherein the value is marginally higher in the case of the aged condition as compared with the homogenized condition, an observation directly attributable to the higher modulus of the
phase as compared to that of the
phase.

7. 5. Summary and conclusions

Two low modulus
-Ti alloys based on the Ti-Nb-Zr-Ta and Ti-Mo-Zr-Fe systems have been investigated in the present study. The microstructure of these alloys in the homogenized and aged conditions have been investigated in detail using SEM and TEM with the objective of identifying the critical strengthening mechanisms operational in these alloys. Results suggest that a metastable B2 ordering in the matrix of the homogenized TNZT alloy results in increased values of both hardness and modulus for this alloy. On ageing, the precipitation of fine-scale secondary
destroys the B2 ordering in the matrix, leading to both softening as well as a decrease in modulus. In contrast, the TMZF alloy exhibits an increase in both hardness and modulus in the aged condition as compared with the homogenized condition. Precipitation of fine-scale secondary
on ageing and the absence of B2 ordering in the matrix results in the observed increase in hardness and modulus.

8. References

1. M.J. Long and H.J. Rack, Titanium alloys in total joint replacement—a materials science perspective. Biomaterials 19 (1998), pp. 1621-1639. Abstract | PDF (341 K)

2. E.W. Lowman, Osteoarthiritis. J Amer Med Acad 157 (1955), p. 487.

3. K. Wang, The use of titanium for medical applications in the USA. Mater Sci Eng A213 (1996), pp. 134-137. Abstract | PDF (325 K)

4. C.M. Lee, W.F. Ho, C.P. Ju and J.H. Chern Lin, Structure and properties of Titanium-25Niobium-xIron alloys. J Mater Sci Mater Med 13 (2002), pp. 695-700. Abstract-EMBASE | Abstract-Compendex   | Full Text via CrossRef

5. M.F. Semlitsch, H. Weber, R.M. Streicher and R. Schön, Joint replacement components made of hot-forged and surface-treated Ti-6Al-7Nb alloy. Biomaterials 13 11 (1992), pp. 781-788. Abstract

6. Borowy K-H, Krammer K-H. On the properties of a new Titanium alloy (Ti-5Al-2.5Fe) as implant material. Titanium 84': Science and Technology, vol. 2. Munich: Desche Ges Metallkunde EV, 1985, p. 1381-6.

7. Steinemann SG. Corrosion of surgical implants—in vivo and in vitro tests. In: Winter GD, Leray JL, de Groot K, editors. Evaluation of biomaterials, New York: Wiley; 1980.

8. Steinemann SG. Corrosion of Titanium, Titanium Alloys for surgical implants. Titanium 84': Science and Technology, vol. 2. Munich: Desche Ges Metallkunde EV, 1985, p. 1373-1379.

9. S. Rao, T. Ushida, T. Tateishi, Y. Okazaki and S. Asao. Bio-med Mater Eng 6 (1996), p. 79. Abstract-MEDLINE  

10. P.R. Walker, J. Leblanc and M. Sikorska. Biochemistry 28 (1990), p. 3911.

11. E. Cheal, M. Spector and W. Hayes, Role of loads and prosthesis material properties on the mechanics of the proximal femur after total hip arthoplasty. J Orthop Res 10 (1992), pp. 405-422. Abstract-Compendex | Abstract-MEDLINE | Abstract-EMBASE  

12. P. Prendergast and D. Taylor, Stress analysis of the proximo-medial femur after total hip replacement. J Biomed Eng 12 5 (1990), pp. 379-382. Abstract-MEDLINE | Abstract-INSPEC | Abstract-EMBASE | Abstract-Compendex  

13. W.F. Ho, C.P. Ju and J.H. Chern Lin, Structure and properties of cast binary Ti-Mo alloys. Biomaterials 20 (1999), pp. 2115-2122. SummaryPlus | Full Text + Links | PDF (1375 K)

14. Wang K, Gustavson L, Dumbleton J. The characterization of Ti-12Mo-6Zr-2Fe. A new biocompatible titanium alloy developed for surgical implants. Beta Titanium in the 1990s. Warrendale, Pennsylvania: The Mineral, Metals and Materials Society; 1993. p. 2697-704.

15. Steinemann SG, Mäusli P-A, Szmukler-Moncler S, Semlitsch M, Pohler O, Hintermann H-E, Perren S-M. Beta-titanium alloy for surgical implants. Beta Titanium in the 1990s. Warrendale, Pennsylvania: The Mineral, Metals and Materials Society; 1993. p. 2689-96.

16. Fanning JC. Properties and processing of a new metastable beta titanium alloy for surgical implant applications: TIMETAL™ 21SRx. Titanium 95': Science and Technology 1996; 1800-07.

17. Mishra AK, Davidson JA, Kovacs P, Poggie RA. Ti-13Nb-13Zr: a new low modulus, high strength, corrosion resistant near-beta alloy for orthopaedic implants. Beta Titanium in the 1990s. Warrendale, Pennsylvania: The Mineral, Metals and Materials Society; 1993. p. 61-72.

18. Ahmed TA, Long M, Silverstri J, Ruiz C, Rack HJ. A new low modulus biocompatible titanium alloy. Titanium 95': Science and Technology 1996; 1760-7.

19. X. Tang, T. Ahmed and H.J. Rack, Phase transformations in Ti-Nb-Ta and Ti-Nb-Ta-Zr alloys. J Mater Sci 35 (2000), pp. 1805-1811. Abstract-Compendex | Abstract-INSPEC   | Full Text via CrossRef

Corresponding author. Tel.: [phone]; fax: [phone]

---