Materials and Design 35 (2012) 731–740

Contents lists available at SciVerse ScienceDirect

Materials and Design

journal homepage: www.elsevier .com/locate /matdes

Technical Report

The influence of chemical composition and thermo-mechanical treatmenton Ti–Nb–Ta–Zr alloys

Jaroslav Málek a,b,⇑, František Hnilica a, Jaroslav Vesely a, Bohumil Smola c, Sonia Bartáková d, Jirí Vanek d

a UJP PRAHA a.s., Nad Kamínkou 1345, 156 00 Praha 5 – Zbraslav, Czech Republicb Czech Technical University in Prague, Faculty of Mechanical Engineering, Karlovo námestí 13, 121 35 Praha 2, Czech Republicc Charles University, Faculty of Mathematics and Physics, Ke Karlovu 5, CZ 121 16 Prague 2, Czech Republicd St. Anne’s University Hospital Brno, Pekarská 53, 656 91 Brno, Czech Republic

a r t i c l e i n f o a b s t r a c t

Article history:Received 13 June 2011Accepted 18 October 2011Available online 28 October 2011

0261-3069/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.matdes.2011.10.030

⇑ Corresponding author at: UJP PRAHA a.s., Nad KamZbraslav, Czech Republic. Tel.: +420 227180380; fax:

E-mail address: malek@ujp.cz (J. Málek).

Ti–Nb–Ta–Zr quaternary alloying system is very promising for biomedical alloys. It is due to goodmechanical properties and corrosion resistance of titanium alloys. Moreover no potentially harmful ele-ments are contained in this system.

Mechanical properties were influenced by changing the chemical composition and by various heat-treatment operations. The alloys were prepared by arc melting and then they were hot forged(900–1000 �C). After solution treatment 850 �C/0.5 h/water quenched, cold swaging was carried out withsection reduction about 85%. As final heat treatment aging at 450 �C/8 h/furnace cooling was used.

Mechanical properties were measured from tensile tests results at cold swaged and aged specimens.The microstructure was observed by using light microscopy and transmission electron microscopy(TEM)-thin foils method. X-ray diffraction analysis reveals the phase composition. By using these tech-niques the changes in microstructure caused by precipitation during aging treatment were clarified. Afteraging, the presence of x or a phases may occur. Influence of changing Zr and Ta contents on mechanicalproperties and also on precipitation of secondary phases during aging treatment was observed.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Titanium has hcp (a-phase) crystallographic structure at roomtemperature that undergoes an allotropic transformation to a bcc(b-phase) at about 883 �C. The temperature of this change can beincreased by adding a-stabilizers (e.g. Al, O, N) or lowered by add-ing b-stabilizers (e.g. Mo, Nb, Ta, V). This temperature is called theb-transus temperature. Titanium and its alloys are nowadays veryfrequently used materials in various applications. This is due totheir good properties (e.g. high strength, low density, good corro-sion resistance). These properties predict them also as materialsfor bio-applications. Due to increasing average age the demandfor biomaterials also increases.

The materials for bio-applications have to meet some specificrequirements. Corrosion resistance, biocompatibility and sufficienttensile strength are essential. Currently used materials (Ti6Al4 V,stainless steel, etc.) contain elements, which can be harmful tohuman body due to the possible release of toxic Al, V or Cr. There-fore only fully biocompatible elements should be used in thesealloys. Ta, Nb, Zr, Hf and Pt are considered as safe elements foralloying in titanium [1–5]. Among alloying systems the ternary

ll rights reserved.

ínkou 1345, 156 00 Praha 5 –+420 227180390.

Ti–Nb–Ta–Zr system is supposed to be suitable for bio-applica-tions. These alloys contain only fully biocompatible elements andmoreover Nb and Ta are b-stabilizers [6–8]. The b-phase has thelowest Young’s modulus among all titanium phases. This is veryimportant for the so called biomechanical compatibility in orderto avoid the ‘‘stress shielding effect’’. This is cause by insufficientloading of the bone, which can cause bone resorption and loosingof the implant [2,9,10]. The addition of Zr improves the strengthof the alloy because of the solution strengthening effect [5,10,11].This is very important for these alloys, because they generally haverelatively low ultimate tensile strength in solution treated condi-tion. Zr is also considered as fully biocompatible element and itinfluences precipitation processes in the alloy and mechanicalproperties. These effects are studied in this paper during thermo-mechanical treatment, which was simulating production of wireswith desired combination of mechanical properties.

2. Methods

The investigated alloys were prepared by arc melting under he-lium atmosphere with non-consumable tungsten electrode. Ingotswere remelted at least six times in order to ensure the chemicalhomogeneity. Alloys with various chemical composition were pre-pared: Ti–35Nb–(10 – x) Ta–xZr, where x was 0; 2.5; 5; 7.5 and10 wt.% (all chemical compositions in this work are in wt.%). Cast

Fig. 1a. Microstructure of hot forged Ti35Nb10Zr alloy (LM – longitudinaldirection).

Fig. 2. Microstructure of hot forged Ti35Nb2.5Ta7.5Zr (LM – longitudinal direction).

732 J. Málek et al. / Materials and Design 35 (2012) 731–740

ingots were machined and then hot forged (900–1000 �C) into acylindrical shape with a diameter about 14 mm (deformationabout 40%). These specimens were then solution treated (850 �C/0.5 h/water quenched). After solution treatment the rods were coldswaged in several steps. Total section reduction was about 85%.Aging (450 �C/8 h/furnace cooled) was used as a final heattreatment.

The microstructure of specimens was studied by using lightmicroscopy (LM), scanning electron microscopy (SEM) with elec-tron back scattered diffraction detector (EBSD) and transmissionelectron microscopy (TEM). For phase identification selected areadiffraction (SAD) has been used. The samples for LM observations(on Nikon EPIPHOT 3000 microscope) were prepared by standardmetallographic processes (grinding, polishing and etching). Crys-tallographic observations were carried out on JEOL JSM 7600Fmicroscope equipped with Nordly’s EBSD detector. The resultswere processed by HKL Channel 5 software equipment. TEM obser-vations of microstructure and phase composition were performed

Fig. 1b. Microstructure of hot forged Ti35Nb1

using JEOL JEM 2000EX transmission electron microscope. Thinfoils for TEM observations were prepared by grinding to a thick-ness of about 100 lm and then electropolished in STRUERS TENU-POL 2 machine by using electrolyte of 10% HClO4, 20% Glycerol and70% methanol at �20 �C. Tensile tests of prepared specimens werecarried out on Instron 1185 machine according to ISO 6892-1:2009standard. Standard round tensile specimens (diameter 4 mm) withround shoulders were used for the test.

3. Results

3.1. Hot forging

The microstructure of as-cast alloys consists of very coarse den-drites (larger than 1 mm). It undergoes significant changes duringhot forging. Some images of hot forged specimens microstructureobtained by LM and by EBSD are shown below (Figs. 1–4). Themicrostructure of hot forged Ti35Nb10Zr alloy consists of very

0Zr alloy (EBSD – longitudinal direction).

Fig. 3a. Microstructure of hot forged Ti35Nb5Ta5Zr alloy (LM – longitudinaldirection).

Fig. 4a. Microstructure of hot forged Ti35Nb10Ta alloy (LM – longitudinal

J. Málek et al. / Materials and Design 35 (2012) 731–740 733

large grains. Within these large grains deformation bands or twinscan be seen. Fine grains having substructure of subgrains are ob-served on the boundaries of these grains (see Fig. 1b). The micro-structure of the Ti35Nb2.5Ta7.5Zr specimen is different. Itconsists of relatively fine grains. In few of them deformation bandsor twins are present (Fig. 2). The Ti35Nb5Ta5Zr specimen exhibitsin general recrystallized grains with deformation bands and smallangle boundaries (see Figs. 3a and 3b). The microstructure of theTi35Nb7.5Ta2.5Zr is similar to that of Ti35Nb5Ta5Zr specimen.Large areas that have similar crystallographic orientation in themicrostructure of the Ti35Nb10Ta specimen can be observed(Fig. 4b). Some low angle boundaries are inside these areas. Nodeformation bands typical for Ti35Nb10Zr were observed. Further-more also fine grains can be found on grain boundaries of largegrains (Fig. 4a). These grains seem to be deformed, but they alsolack deformation bands. Only low angle boundaries were detectedin their interior.

Fig. 3b. Microstructure of hot forged Ti35Nb5T

3.2. Solution treatment

The microstructure of the solution treated specimens consists ofb-phase and is presented in Figs. 5–7. The microstructure of theTi35Nb10Zr alloy after solution treatment consists of relatively fineequiaxed grains. Larger grains are present only locally (see Figs. 5aand 5b). No subgrains were observed in Ti35Nb10Zr alloy. Solutiontreatment has no significant effect on the microstructure of theTi35Nb5Ta5Zr alloy (see Figs. 6a and 6b). Large grains slightly elon-gated in one direction can be found there, but a few small angleboundaries within the grains are observed (this microstructure issimilar also for Ti35Nb7.5Ta2.5Zr specimen). Microstructure ofthe solution treated Ti35Nb10Ta alloy consists of equiaxed grains.These grains are coarser than those in solution treated Ti35Nb10Zralloy. They are deformed and small angle boundaries can be seenthere (see Fig. 7b).

a5Zr alloy (EBSD – longitudinal direction).

direction).

Fig. 4b. Microstructure of hot forged Ti35Nb10Ta alloy (EBSD – longitudinal direction).

Fig. 5a. Microstructure of solution treated Ti35Nb10Zr alloy (LM – longitudinaldirection).

734 J. Málek et al. / Materials and Design 35 (2012) 731–740

3.3. Cold swaging

The specimens are highly deformed during cold swaging. Be-cause of that the EBSD analysis was difficult and obtained resultsare not reliable (they are not presented here). The grains are elon-gated in one direction (Figs. 8 and 9). TEM observations revealed b-phase in all studied specimens. Moreover a0 0-phase was clearlyidentified in SAD of cold swaged alloys with high Zr content –Ti35Nb10Zr and Ti35Nb2.5Ta7.5Zr (see Fig. 10 – Ti35Nb2.5-Ta7.5Zr). Only b-phase grains and dislocation tangles were recog-nized in the Ti35Nb5Ta5Zr (see Fig. 11) and also inTi35Nb7.5Ta2.5Zr and in Ti35Nb10Ta.

3.4. Aging

High volume fraction of needle like particles precipitated in theTi35Nb10Zr alloy during aging treatment (Fig. 12). They were iden-tified as a-phase precipitates. Also x-phase and a0 0-phase spots

were identified locally in SAD, however these spots were weak.So it can be concluded, that the volume fraction of a0 0 and x phasesis relatively low. No precipitates were detected in the bright fieldimage (BF) of aged Ti35Nb5Ta5Zr (Fig. 13), but reflections of x-precipitates were identified in SAD. So the volume fraction of theseprecipitates is relatively low. The a-phase in Ti35Nb10Ta precipi-tated in the form of needle-like particles – bright needles in thedark field (DF) micrograph and was identified in SAD (Fig. 14).The volume fraction of a-phase needles in Ti35Nb10Ta was lowerthan that in Ti35Nb10Zr alloy. In some places also weak spots of x-phase were observed.

3.5. Mechanical properties

Results of tensile tests are presented in Figs. 15a–15d. Tensilestrength (Rm) of the cold swaged Ti35Nb10Zr specimen is the high-est among all cold swaged alloys. It decreases with decreasing Zrcontent (and increasing Ta content). The tensile strength of theTi35Nb2.5Ta7.5Zr, Ti35Nb5Ta5Zr and Ti35Nb7.5Ta2.5Zr specimensis similar and within experimental scatter does not depend on Zr orTa content. The tensile strength of the Ti35Nb10Ta is higher thanfor above mentioned specimens but is not as high as that of theTi35Nb10Zr alloy (see Fig. 15a). The values of the 0.2 proof strength(RP02) follow the same trend (Fig. 15b). The elongation (A5) valuesof cold swaged specimens are similar within experimental scatter(see Fig. 15c). The value of Young’s modulus (E) reaches the maxi-mum for Ti35Nb5Ta5Zr alloy and falls down towards the values foralloys with 10% Zr or 10% Ta. The lowest value (50 GPa) has the for-mer one (Fig. 15d). With respect to experimental scatter all valuesof Young’s modulus, except that of Ti35Nb10Zr are similar.

Increase in tensile strength values is evident for all observed al-loys after aging treatment. It is most distinct in the Ti35Nb10Zr al-loy. The increase for all other alloys is similar. The 0.2 proofstrength values exhibit similar trend. Elongation decreases signifi-cantly for alloy with 10% Zr, but for other alloys has similar valuesas before aging treatment. The Young’s modulus has increased inall studied alloys during aging treatment. This is most significantin Ti35Nb10Zr alloy which has the highest modulus after aging.Than the modulus values have decreasing trend with decreasing

Fig. 5b. Microstructure of solution treated Ti35Nb10Zr alloy (EBSD – longitudinal direction).

Fig. 6a. Microstructure of solution treated Ti35Nb5Ta5Zr alloy (LM – longitudinaldirection).

J. Málek et al. / Materials and Design 35 (2012) 731–740 735

Zr content in the alloy, but the differences are small (within exper-imental scatter).

4. Discussion

The b-transus temperature for studied alloys is about 600 �C(according to [12]). So the hot forging process (deformation is40% for all the specimens) is carried out in b-phase field. Duringhot forging dynamical recovery and dynamical recrystallizationtake place. Deformed large grains with deformation bands arepresent in the Ti35Nb10Zr alloy (Fig. 1b). Recrystallized grainswith subgrains are observed at boundaries of large grains. Thisobservation is consistent with Weiss and Semiatin [12] and withFuruhara et al. [13], who reported, that dynamical recrystallizationmay occur preferentially in areas with higher deformation (e.g.grain boundaries). High angle boundaries are developed due tocoalescence of small angle boundaries during recrystallization pro-

cess [12]. This recrystallization process can be, most probably con-sidered as the one that takes place in the Ti35Nb10Zr andTi35Nb10Ta alloys during hot forging. Ti35Nb10Ta exhibits largegrains containing small angle boundaries and locally fine recrystal-lized grains on boundaries of the large ones (Figs. 4a and 4b). Onthe other hand Ti35Nb5Ta5Zr is recrystallized after hot forging.Some grains are slightly deformed and rarely some subgrains oc-curred (Fig. 3b). More detailed explanation of the influence ofchemical composition on recrystallization needs further study.

After solution treatment, the microstructure of the Ti35Nb10Zrconsists of relatively fine equiaxed grains and locally also largergrains can be found. Fine grains are formed in areas, where largedeformed grains were present before solution treatment. In thesehighly deformed areas enough deformation energy is stored forpost-dynamic recrystallization process, which is in accordancewith Weiss and Semiatin [12]. The Ti35Nb5Ta5Zr exhibits alsorecrystallized microstructure after solution treatment with grainselongated in one direction. The Ti35Nb5Ta5Zr alloy was recrystal-lized after hot forging and the grains were slightly deformed. Nosignificant changes in microstructure were observed (only fewsubgrains disappeared), because of the lack of deformation energyfor recrystallization during solution treatment and therefore onlydynamic recovery takes place in this specimen. Ti35Nb10Ta isnot fully recrystallized after solution treatment. Subgrains withinrelatively large equiaxed grains can be seen in the microstructure(EBSD image).

After cold swaging, no other phases than b-phase and a0 0-phasewere observed by TEM. The a0 0-phase was observed in alloys withhigher Zr content (10% and 7.5% – Fig. 10). This can be explained bystress induced martensitic transformation (SIM) during the coldswaging. The occurence of SIM is due to lower b-phase stabilityin these alloys after solution treatment (Ti35Nb10Zr andTi35Nb2.5Ta7.5Zr) [3,14,15]. Other alloys have probably higherb-phase stability and therefore no (or less) SIM takes place there.The stability of b-phase can be evaluated by considering the Moeqv

[13]. The higher the Moeqv, the higher is the b-phase stability. ForTi35Nb10Zr the Moeqv is 9.8 and it increases with increasing Tacontent to 11.8 for Ti35Nb10Ta alloy. It should be pointed out, that

Fig. 6b. Microstructure of solution treated Ti35Nb5Ta5Zr alloy (EBSD – longitudinal direction).

Fig. 7a. Microstructure of solution treated Ti35Nb10Ta alloy (LM – longitudinaldirection).

736 J. Málek et al. / Materials and Design 35 (2012) 731–740

the influence of Zr is not considered in Moeqv. Zr enhances the sta-bilizing effect of Nb and Ta, because it hinders the transformationa M b. It was reported by Ferrandini et al. [10] that the b-stabiliz-ing effect of Zr in these alloys decreases with its increasing content.

The influence of Zr on b-phase stability is involved in DVX-acluster method, where electronic parameters Bo (bond order), Md

(d-orbital energy level) and e/a (electron/atom ratio) are used[16–18]. These parameters are calculated for each alloy asBo ¼

PðXi � BOiÞ and Md ¼

PðXi �MDiÞwhere Xi is the molar fraction

of the i element (i = Ti, Nb, Ta, Zr) and BOi and MDi are values of BO

and MD for this element [19]. Bo and Md values give the Bo �Md

map where a, a0 0, b + a, b + x and b-phase regions are defined foralloys in as-quenched state. Bo and Md values for studied alloyswere calculated and can be found in Table 1.

Ti35Nb10Ta and Ti35Nb7.5Ta2.5Zr are in single b-phase regionaccording to their Bo, Md values. On the other hand Ti35Nb2.5-Ta7.5Zr and Ti35Nb10Zr are in area where b + a or b + x phases

coexist. So they have lower b-phase stability in as-quenched condi-tion than Ti35Nb10Ta and Ti35Nb7.5Ta2.5Zr. Additional informa-tion about as-quenched microstructure can be obtained from e/avalues. With increasing e/a ratio the b-phase stability is higher.According to [16] the as-quenched alloy has microstructure con-sisting of single b-phase for e/a values higher than 4.20. All studiedalloys have higher e/a values, but Ti35Nb10Zr has the lowest valueof e/a which means that in this alloy the b-phase stability is lowand SIM can occured easily during deformation. Hwang et al.[17] have shown the relationship between plastic deformationbehavior and Bo and e/a values for Ti–Nb–Ta–Zr type alloys. SIMtakes place in alloys with lower Bo and e/a values. On the otherhand dislocation slip occurs for higher Bo and e/a values. Alloyshaving Bo � 2:87 and e/a � 4.24 exhibit peculiar plastic deforma-tion (plastic deformation without any dislocation activity) [17].Our results confirmed SIM occurrence for lower Bo and e/a valuesand dislocation slip for higher Boand e/a values, but no peculiarplastic deformation were observed in Ti35Nb10Zr alloy(Bo ¼ 2:87 and e/a = 4.24.).

After aging treatment the a-phase was observed in Ti35Nb10Zr(Fig. 12) and in Ti35Nb10Ta (Fig. 14) alloys. Detailed study of agedspecimens by TEM revealed the presence of certain fraction of a-phase and very low fraction of x-phase in Ti35Nb10Ta. InTi35Nb5Ta5Zr the spots on SAD resulting from x-phase presencewere clearly identified (Fig. 13). In Ti35Nb10Zr the a-phase wasclearly identified and very weak spots of x-phase were also foundin SAD. Hao et al. [20] reported that during aging treatment x-phase occurs at lower aging temperatures (300–400 �C). Longeraging times at medium temperatures (400–450 �C) favor a-phaseprecipitation instead of x-phase, that precipitates at these temper-atures after shorter aging periods. Similar results were obtained byvarious authors (e.g. [2,7,21–23]). At higher aging temperatures(450–550 �C) a-phase precipitated. The a-phase nucleates on pre-viously precipitated x-phase particles [24,25]. The more stable isthe b-phase, the higher temperature or the longer aging time isneeded for a-phase precipitation. The changes during aging weredescribed in the following transformation sequence published in[23] by Afonso et al.

Fig. 7b. Microstructure of solution treated Ti35Nb10Ta alloy (EBSD – longitudinal direction).

Fig. 8. Microstructure of cold swaged Ti35Nb10Ta alloy (LM – longitudinaldirection).

Fig. 9. Microstructure of cold swaged Ti35Nb10Zr alloy (LM – transverse direction).

Fig. 10. Microstructure of cold swaged Ti35Nb2.5Ta7.5Zr alloy with SAD (TEM-BF).

J. Málek et al. / Materials and Design 35 (2012) 731–740 737

b! bþ b0 ! bþx! bþ a

where b0 is tetragonal structure (bct) similar to b-phase.The absence of Ta implies fast transformation (lower b-phase

stability) in Ti35Nb10Zr alloy where only a-phase was proved. InTi35Nb10Ta the stability of b-phase is higher than in Ti35Nb10Zr.The presence of x-phase in this alloy was also observed but not asclearly as in Ti35Nb5Ta5Zr. The precipitation observed duringaging is in agreement with the transformation sequence, butTi35Nb10Zr alloy exhibits higher volume fraction of a-phase afteraging than Ti35Nb10Ta. The transformation to a-phase was notcompleted in Ti35Nb5Ta5Zr and only x-phase precipitated. So

Fig. 11. Microstructure of cold swaged Ti35Nb5Ta5Zr alloy with SAD (TEM-BF).

Fig. 12. Microstructure of aged Ti35Nb10Zr alloy with SAD – b matrix and a-phaseprecipitates (TEM-BF).

Fig. 13. Microstructure of aged Ti35Nb5Ta5Zr alloy with SAD (TEM-BF).

Fig. 14. Microstructure of aged Ti35Nb10Ta alloy with SAD – b matrix and a-phaseprecipitates (TEM-DF).

738 J. Málek et al. / Materials and Design 35 (2012) 731–740

Ti35Nb5Ta5Zr is supposed to have the highest b-phase stability be-fore aging among studied alloys according to the transformationsequence. On the other hand this is not consistent with the abovediscussed theories (Moeqv and Bo, Md). It should be mentioned thatalso deformation microstructure can affect precipitation processes.

Phase composition and stability of b-phase influence mechani-cal properties of the alloys. In cold swaged condition theTi35Nb10Zr alloy exhibits the highest value of tensile strengthamong all studied cold swaged alloys. This can be ascribed tostrong solid solution strengthening effect of Zr [5,10,26] and par-tially to strengthening effect of dispersed a0 0 phase, but this effectis weak [27,28]. The tensile strength values of alloys with 2.5%, 5%and 7.5% Zr are similar (Fig. 15a). Slightly higher value of tensile

strength can be observed in Ti35Nb10Ta. This can be ascribed tostrengthening effect of Ta [29]. This trend was proved in the 0.2proof strength values.

After aging treatment the increase in tensile and yield strengthis most significant in Ti35Nb10Zr alloy. It is connected with highvolume fraction of a-phase precipitated during aging treatment.Also the SIM deformation mechanism is less probable after agingbecause of the stabilization of b-phase during aging treatment.The b-phase is stabilized during aging through transformation tox or/and a-phases (these precipitates contain less b-stabilizersthan b-phase). Therefore the solid solution strengthening effectof Zr can become more significant. Also in alloys with lower Zr con-tent a and x-phases caused increase in strength. Their strengthen-ing effect depends on their dimensions, morphology, coherency,

Fig. 15a. Tensile strength (Rm) of alloys in cold swaged and in aged condition.

Fig. 15b. 0.2 Proof stress (Rp02) of alloys in cold swaged and in aged condition.

Fig. 15c. Elongation (A5) of alloys in cold swaged and in aged condition.

Fig. 15d. Young’s modulus (E) of alloys in cold swaged and in aged condition.

Table 1Calculated Bo , Md and e/a values for studied alloys.

Ti35Nb10Zr

Ti35Nb2.5Ta7.5Zr

Ti35Nb5Ta5Zr

Ti35Nb7.5Ta2.5Zr

Ti35Nb10Ta

Bo 2.870 2.869 2.868 2.866 2.865

Md 2.465 2.458 2.450 2.443 2.435e/a 4.24 4.25 4.26 4.27 4.28

J. Málek et al. / Materials and Design 35 (2012) 731–740 739

and volume fraction. The 0.2 proof strength follows the same trendas that of the tensile strength. The elongation values are more orless consistent with these results. Ti35Nb10Zr has the lowest valueof elongation in aged condition and other samples have similar val-ues varying within 2%.

The Young’s modulus after cold swaging reaches the maximumfor Ti35Nb5Ta5Zr alloy and then decreases with increasing Zr con-tent and exhibits minimum for Ti35Nb10Zr (50 GPa). This is prob-ably caused due to higher Zr presence, which causes an increase inlattice parameters and therefore the distance between atoms are

higher and that results in weaker bonding and lower Young’s mod-ulus [30]. The Ta addition can cause an increase in modulus due tohigher modulus of pure Ta (in comparison with Ti) as was reportedby Elias et al. [29]. The results in this case are not clear due to smalldifferences between the values of modulus and relatively highexperimental scatter. Hwang et al. [17] have shown that there iscorrelation between Young’s modulus and Bo values for as-quenched Ti–Nb–Ta–Zr type alloys. On the other hand this correla-tion was not found for Ti–Nb–Ta–Zr alloys after deformation whichis consistent with our results.

The aging treatment leads to an increase in modulus of all stud-ied alloys. Young’s modulus of aged alloys tends to decrease fromthe highest value of 90 GPa (Ti35Nb10Zr) to about 77 GPa(Ti35Nb10Ta) after aging treatment. This is caused by differentphase composition and volume fraction of phases as was reportedby Majumdar et al. [31]. Ti35Nb10Zr exhibits the highest Young’smodulus among all aged alloys resulting from the highest volumefraction of a-phase with high Young’s modulus in microstructure.High volume fraction of this phase also causes high increase in ten-sile strength and 0.2 proof strength values. Also the modulus in-crease during aging treatment was the highest in this alloy.Ti35Nb5Ta5Zr has low fraction of x-phase. This phase has thehighest modulus among all phases in titanium alloys [31], butthe fraction of x-phase is much lower in this alloy than that ofa-phase in Ti35Nb10Zr. This causes the lower modulus in compar-ison with the Ti35Nb10Zr alloy. Also the tensile strength and 0.2proof strength values increased less than in Ti35Nb10Zr alloy.Ti35Nb10Ta after aging treatment has lower volume fraction ofa-phase than the Ti35Nb10Zr alloy. This results in lower increasein Young’s modulus and strength characteristics in comparisonwith Ti35Nb10Zr alloy. It can be concluded that Young’s modulus,tensile strength and 0.2 proof strength values are increased due toprecipitation of a or x-phases, which is consistent with [20,22].The value of that increase depends on their volume fraction anddensity. These processes can be controlled through chemical com-position (b-phase stability) and heat treatment.

740 J. Málek et al. / Materials and Design 35 (2012) 731–740

5. Conclusions

The microstructure and mechanical properties of alloys withchanging chemical composition were studied during thermo-mechanical processes. On the basis of obtained results can beconcluded:

a) The change of Ta and Zr content leads to different mecha-nisms of dynamical recovery and dynamical recrystallizationduring hot forging of as-cast alloys. The Ti35Nb5TaZr alloywas dynamically recrystallized, but Ti35Nb10Ta andTi35Nb10Zr alloys were recrystallized only locally on grainboundaries of original grains.

b) After solution treatment (850 �C/0.5 h), the microstructureof all alloys consist of recrystallized grains formed by post-dynamic recrystallization or dynamic recrystallization dur-ing previous hot forging (Ti35Nb5Ta5Zr).

c) After cold swaging the specimens have deformed micro-structure with grains elongated in rod axis direction. Thealloys with higher Zr content (7.5% and 10%) exhibit a0 0 mar-tensitic phase in the b-matrix.

d) Needle like particles were observed in microstructure ofTi35Nb10Ta and Ti35Nb10Zr alloys after aging treatment(450 �C/8 h). In Ti35Nb5Ta5Zr alloy x-precipitates arepresent.

e) Ti35Nb10Zr posses the highest tensile strength and 0.2 proofstrength in both cold swaged and aged states. The differ-ences among other alloys are not so evident.

f) Young’s modulus of these alloys exhibits values from 50 to65 GPa in cold swaged state and 75 to 90 GPa after agingtreatment.

Acknowledgements

This work was supported by Department of Commerce of CzechRepublic – research program ‘‘Trvala prosperita’’ No. 2A-2TP1/073.

References

[1] Wang L, Lu W, Qin J, Zhang F, Zhang D. Microstructure and mechanicalproperties of cold-rolled TiNbTaZr biomedical b titanium alloy. Mater Sci Eng:A 2008;490:421.

[2] Wang L, Lu W, Qin J, Zhang F, Zhang D. Effect of precipitation phase onmicrostructure and superelasticity of cold-rolled beta titanium alloy duringheat treatment. Mater Des 2009;30:3873.

[3] Niinomi M, Akahori T, Nakai M. In situ X-ray analysis of mechanism ofnonlinear super elastic behavior of Ti–Nb–Ta–Zr system beta-type titaniumalloy for biomedical applications. Mater Sci Eng: C 2008;28:406.

[4] Souza SA, Manicardi RB, Ferrandini PL, Afonso CRM, Ramirez AJ, Caram R. Effectof the addition of Ta on microstructure and properties of Ti–Nb alloys. J AlloysCompd 2010;504:330.

[5] Ribeiro A, Junior R, Cardoso F, Filho R, Vaz L. Mechanical, physical, andchemical characterization of Ti–35Nb–5Zr and Ti–35Nb–10Zr casting alloys. JMater Sci Mater Med 2009;20:1629–36.

[6] Hou FQ, Li SJ, Hao YL, Yang R. Nonlinear elastic deformation behaviour of Ti–30Nb–12Zr alloys. Scr Mater 2010;63:54.

[7] Ikeda M, Komatsu S, Sowa I, Niinomi M. Aging behavior of the Ti–29Nb–13Ta–4.6Zr new beta alloy for medical implants. Metall Mater Trans A2002;33:487–93.

[8] Sander B, Raabe D. Texture inhomogeneity in a Ti–Nb-based b-titanium alloyafter warm rolling and recrystallization. Mater Sci Eng: A 2008;479:236.

[9] Wang L, Lu W, Qin J, Zhang F, Zhang D. The characterization of shape memoryeffect for low elastic modulus biomedical [beta]-type titanium alloy. MaterCharact 2010;61:535.

[10] Ferrandini PL, Cardoso FF, Souza SA, Afonso CR, Caram R. Aging response of theTi–35Nb–7Zr–5Ta and Ti–35Nb–7Ta alloys. J Alloys Compd 2007;433:207.

[11] Martins DQ, Osorio WR, Souza MEP, Caram R, Garcia A. Effects of Zr content onmicrostructure and corrosion resistance of Ti–30Nb–Zr casting alloys forbiomedical applications. Electrochim Acta 2008;53:2809.

[12] Weiss I, Semiatin SL. Thermomechanical processing of beta titanium alloys –an overview. Mater Sci Eng A 1998;243:46.

[13] Furuhara T, Poorganji B, Abe H, Maki T. Dynamic recovery and recrystallizationin titanium alloys by hot deformation. JOM J Miner, Met Mater Soc2007;59:64–7.

[14] Kuramoto S, Furuta T, Hwang J, Nishino K, Saito T. Elastic properties of gummetal. Mater Sci Eng: A 2006;442:454.

[15] Eckert J, Das J, Xu W, Theissmann R. Nanoscale mechanism and intrinsicstructure related deformation of Ti-alloys. Mater Sci Eng: A 2008;493:71.

[16] Laheurte P, Prima F, Eberhardt A, Gloriant T, Wary M, Patoor E. Mechanicalproperties of low modulus b titanium alloys designed from the electronicapproach. J Mech Behav Biomed Mater 2010;3:565.

[17] Hwang J, Kuramoto S, Furuta T, Nishino K, Saito T. Phase-stability dependenceof plastic deformation behavior in Ti–Nb–Ta–Zr–O alloys. J Mater Eng Perform2005;14:747–54.

[18] Kuroda D, Niinomi M, Morinaga M, Kato Y, Yashiro T. Design and mechanicalproperties of new b type titanium alloys for implant materials. Mater Sci Eng A1998;243:244.

[19] Abdel-Hady M, Hinoshita K, Morinaga M. General approach to phase stabilityand elastic properties of b-type Ti-alloys using electronic parameters. ScrMater 2006;55:477.

[20] Hao Y, Yang R, Niinomi M, Kuroda D, Zhou Y, Fukunaga K, et al. Aging responseof the young’s modulus and mechanical properties of Ti–29Nb–13Ta–4.6Zr forbiomedical applications. Metall Mater Trans A 2003;34:1007–12.

[21] Qazi JI, Marquardt B, Allard LF, Rack HJ. Phase transformations in Ti–35Nb–7Zr–5Ta–(0.06–0.68)O alloys. Mater Sci Eng: C 2005;25:389.

[22] Zhou YL, Niinomi M, Akahori T. Decomposition of martensite a0 0 during agingtreatments and resulting mechanical properties of Ti–Ta alloys. Mater Sci EngA 2004;384:92.

[23] Afonso CRM, Ferrandini PL, Ramirez AJ, Caram R. High resolution transmissionelectron microscopy study of the hardening mechanism through phaseseparation in a b-Ti–35Nb–7Zr–5Ta alloy for implant applications. ActaBiomater 2010;6:1625.

[24] Ohmori Y, Ogo T, Nakai K, Kobayashi S. Effects of x-phase precipitation onb! a;a00 transformations in a metastable b titanium alloy. Mater Sci Eng A2001;312:182.

[25] Prima F, Vermaut P, Texier G, Ansel D, Gloriant T. Evidence of a-nanophaseheterogeneous nucleation from x particles in a b-metastable Ti-based alloy byhigh-resolution electron microscopy. Scr Mater 2006;54:645.

[26] Nakasuji K, Okada M. New high strength titanium alloy Ti10%Zr for spectacleframes. Mater Sci Eng: A 1996;213:162–5.

[27] Zhou Y, Niinomi M. Microstructures and mechanical properties of Ti–50 mass%Ta alloy for biomedical applications. J Alloys Compd 2008;466:535.

[28] Cui Y, Li Y, Luo K, Xu H. Microstructure and shape memory effect of Ti–20Zr–10Nb alloy. Mater Sci Eng: A 2010;527:652.

[29] Elias LM, Schneider SG, Schneider S, Silva HM, Malvisi F. Microstructural andmechanical characterization of biomedical Ti–Nb–Zr(–Ta) alloys. Mater SciEng: A 2006;432:108.

[30] Song Y, Xu DS, Yang R, Li D, Wu WT, Guo ZX. Theoretical study of the effects ofalloying elements on the strength and modulus of b-type bio-titanium alloys.Mater Sci Eng: A 1999;260:269.

[31] Majumdar P, Singh SB, Chakraborty M. Elastic modulus of biomedical titaniumalloys by nano-indentation and ultrasonic techniques – a comparative study.Mater Sci Eng: A 2008;489:419.