Microhardness and microstructure evolution of ultra

0 downloads 0 Views 2MB Size Report
Nov 12, 2016 - UFG α+β Ti alloys such as Ti-6Al-4 V alloy and specialized biocompa- ...... [54] C.S. Meredith, A.S. Khan, The microstructural evolution and ...
Materials Science & Engineering A 682 (2017) 220–228

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

crossmark

Microhardness and microstructure evolution of ultra-fine grained Ti-15Mo and TIMETAL LCB alloys prepared by high pressure torsion ⁎

Kristína Václavováa, , Josef Stráskýa, Veronika Polyakovab, Jitka Stráskáa, Jitka Nejezchlebovác, Hanuš Seinerd, Irina Semenovab, Miloš Janečeka a

Charles University in Prague, Department of Physics of Materials, Ke Karlovu 5, Prague 121 16, Czech Republic UFA State Aviation Technical University, Institute of Physics of Advanced Materials, K. Marx Street 12, Ufa 450 000, Russia c Czech Technical University in Prague, Faculty of Nuclear Sciences and Physical Engineering, Trojanova 13, Prague 120 00, Czech Republic d Academy of Sciences of the Czech Republic, Institute of Thermomechanics, Dolejškova 5, 18 200 Prague, Czech Republic b

A R T I C L E I N F O

A BS T RAC T

Keywords: Metastable β-Ti alloys High pressure torsion Microhardness Microstructure evolution Twinning-induced grain refinement Decreasing elastic constants

Two metastable β-Ti alloys, Ti-15Mo and Ti-6.8Mo-4.5Fe-1.5Al (TIMETAL LCB) were solution treated and subjected to severe plastic deformation by high pressure torsion. The evolution of microhardness, microstructure and elastic constants with increasing strain imposed by high pressure torsion was investigated. Fragmentation of the microstructure with increasing strain was observed by scanning electron microscopy. Significant twinning in system {1 1 2} < 111 > after high pressure torsion was observed in both studied alloys by electron backscatter diffraction. Multiple twinning contributes significantly to the fragmentation of grains and consequently to the overall refinement of the microstructure. Microhardness significantly increases with increasing strain and was fitted using the Hollomon and Voce laws. Hollomon's hardenability exponent is much higher for both studied β-Ti alloys than for the commonly used Ti-6Al-4V alloy. It reflects high capability of strengthening β-Ti alloys by intensive plastic deformation. The measurement of elastic constants using resonant ultrasound spectroscopy showed that the deformation by high pressure torsion increases the Young's modulus as compared to solution treated material. On the other hand, further straining causes subsequent decrease of the Young's modulus.

1. Introduction The importance of the β-titanium alloys in commercial practice has been increasing in the last few decades due to successful utilizing their unique properties such as high strength, low specific density, strengthening capability, high fracture toughness, and excellent corrosion resistance [1,2]. β-Ti alloys are extensively used in aircraft industry [3] and considered as prospective candidates for biomedical implants manufacturing due to their excellent biocompatibility and relatively low Young's modulus preventing the stress shielding [4–8]. However, a high strength condition is usually achieved by advanced thermomechanical treatment involving precipitation of α-phase particles, which significantly increases the Young's modulus [9,10]. Severe plastic deformation (SPD) methods strengthen metallic materials via reducing the grain size and increasing the dislocation density [11]. Furthermore, Young's modulus can be reduced by the microstructural refinement. It was also reported that the functional properties such as corrosion resistance and biocompatibility might be also improved by the microstructure refinement [12]. ⁎

Ultra-fine grained (UFG) commercially pure titanium (CP Ti) was prepared by high pressure torsion (HPT) [13] and equal-channel angular pressing (ECAP) [14] almost two decades ago. Furthermore, UFG α+β Ti alloys such as Ti-6Al-4 V alloy and specialized biocompatible Ti-6Al-7Nb alloy were also studied in detail [15,16] and exhibited significantly improved strength and fatigue resistance [17,18]. On the other hand, there is only limited literature on the UFG metastable β-Ti alloys. Reports focused primarily on the study of the enhanced strength, fatigue performance [19,20] and microstructural refinement [21,22]. The elastic properties of UFG β-Ti alloy were studied only in alloys containing niobium as the main alloying element [23,24]. The mechanism of the grain refinement in the β-titanium alloys can vary widely, depending on the specific alloy composition, grain size, deformation mode, temperature and pressure [25]. The classical mechanism of the grain refinement is based on the movement of dislocations, formation of dislocation walls and sub-grain boundaries followed by lattice rotation forming high-angle grain boundaries [26]. However, the grain refinement can be also induced by twinning

Corresponding author. E-mail address: [email protected] (K. Václavová).

http://dx.doi.org/10.1016/j.msea.2016.11.038 Received 18 July 2016; Received in revised form 25 October 2016; Accepted 11 November 2016 Available online 12 November 2016 0921-5093/ © 2016 Elsevier B.V. All rights reserved.

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Fig. 1. Microhardness of Ti-15Mo alloy after various numbers of HPT turns represented by color-coded polar diagram. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

followed by lattice rotation. The two main mechanical twinning systems in bcc materials including β-Ti alloys are {1 1 2}〈1 1 1〉 [27–30] and {3 3 2}〈1 1 3〉[31,32], the latter one being activated especially at low temperatures and at high strain rates. Two metastable β-Ti alloys were used in this study: Ti-15Mo alloy and Ti-6.8Mo-4.5Fe-1.5Al alloy (TIMETAL LCB). Ti-15Mo alloy exhibits good mechanical properties, does not contain any toxic elements such as vanadium and therefore is suitable for a medical use. The TIMETAL LCB (low-cost beta) alloy is characterized by lower production costs due to the partial replacement of the relatively expensive β stabilizing alloying elements such as vanadium or molybdenum by iron. The TIMETAL LCB alloy was commercially used for suspension springs manufacturing, since the weight of a suspension spring can be reduced by 60% thanks to the high strength, low density and reduced Young's modulus of the alloy [9] when compared to previously used steels. The objective of the present study is to examine the evolution of the microstructure, mechanical and elastic properties of two metastable βTi alloys with imposed equivalent strain.

εvonMises=

2πNr , 3h

(1)

where N is the number of rotations, r represents the distance from the sample centre and h is the final thickness of the specimen. The equivalent strain imposed by pressing is about 1.5, while equivalent strains imposed by the torsion are by two decades higher. The thickness is therefore neglected. Microhardness measurements were carried out using the automatic microhardness tester Qness Q10a by Vickers method; 1 kg load and dwell time of 10 s were applied. More than 1000 indents were automatically evaluated along concentric circles, which allows a detailed investigation of microhardness variations on both the surface and the cross-section of the disc. The scanning electron microscope FEI Quanta 200 FX operated at 10 kV was used for microstructural observations and electron backscatter diffraction (EBSD) analysis. Young's modulus and Poisson's ratio were evaluated by the resonant ultrasound spectroscopy (RUS) [35] using a fully contactless laser-based RUS set-up described in detail in [36]. This set-up utilizes focused laser pulses for generating the vibrations in the examined sample and the scanning laser beam for the interferometric detection of the modal response. Five different conditions after different stages of straining were used for RUS measurements for each alloy. All samples were rectangular parallelepipeds with the approximate dimensions of 2×2×1 mm3. For the RUS measurements, the materials of all samples were considered as elastically isotropic, with only two independent elastic constants: Youngʼs modulus E and Poissonʼs ratio ν.

2. Experimental Ti-15Mo alloy was supplied by Carpenter Technology Corp. in a form of a rod with the diameter of 10.5 mm. Ti-6.8Mo-4.5Fe-1.5Al alloy was produced on demand by Huizhou Top Metals Ltd. using magnetic levitation melting and finally wire-cut to the diameter of 20 mm. The as-delivered material was solution treated (1083 K, 20 min) in a protective Ar atmosphere and water quenched. Ti-15Mo alloy was further cut to cylinders (diameter 10.5 mm, height approx. 5 mm) and pressed in HPT machine at room temperature to obtain the desired diameter of 20 mm. The principle of HPT method is described in detail in [33]. Samples with the diameter of 20 mm and the thickness of 1 mm were prepared by HPT at Ufa State Aviation Technical University (USATU) Ufa, Russia at room temperature and the pressure of 2 GPa. A series of samples after N=¼, ½, 1, 3, 5 and 10 turns and N =¼, ½, 1 and 5 turns of HPT was prepared from Ti-15Mo alloy and TIMETAL LCB alloy, respectively. The total equivalent strain imposed in the sample by HPT can be expressed by the von Mises approach, which utilizes a simple torsion, and the strain is then expressed by the linear relation [34]:

3. Results 3.1. Microhardness 3.1.1. Ti-15Mo The microhardness evolution with the increasing number of HPT turns on the specimen's surface is depicted as a series of color-coded images in the Fig. 1. The variations of the microhardness in the crosssection of the specimens are shown in the Fig. 2. The microhardness increases with the increasing distance r from the centre and with the increasing number N of HPT turns. In each image in the Fig. 1. two distinct regions are clearly visible – a central region with a low 221

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Fig. 2. Microhardness variations through thickness of Ti-15Mo alloy processed by HPT (vertical axis is extended for better visibility). (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

15Mo alloy. Vickers microhardness of the TIMETAL LCB alloy after N =¼, ½, 1 and 5 HPT turns was measured on the sample surface (Fig. 3.) and in the cross-section (Fig. 4). Similarly as in the Ti-15Mo alloy, the microhardness increases with the number of HPT turns and also with the increasing distance from the centre of the specimen with exception of specimen edge, where the decreased microhardness is caused by the material outflow. The microhardness in TIMETAL LCB is more homogenous than in in the case of Ti-15Mo, especially after 5 rotations.

microhardness (blue color) and the peripheral region with an increased microhardness (red color). The microhardness in the peripheral region increases with the increasing number of HPT turns. The decrease of the microhardness near the edge of the sample is caused by the material outflow during HPT processing. The microhardness is not homogeneous through the sample thickness (Fig. 2). However, no systematic differences were identified. 3.1.2. TIMETAL LCB The microhardness evolution in the TIMETAL LCB alloy after different number of HPT turns is presented similarly as for the Ti-

Fig. 3. Microhardness of TIMETAL LCB alloy after various number of HPT turns represented on color-coded polar diagram. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

222

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Fig. 4. Microhardness variations through thickness of TIMETAL LCB alloy processed by HPT (vertical axis is extended for better visibility). (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

3.2. Microstructure evolution/refinement

Table 1 EDX analysis of bands occurring in the periphery of the sample.

Scanning electron microscopy was used to study initial stages of the microstructure refinement in both alloys during HPT. Due to the expected lateral inhomogeneity of the microstructure caused by an inhomogeneous character of the imposed strain by HPT, the microstructure was investigated in three regions, namely in the centre of the disc, near the half of the radius of the disc (referred to as “middle”) and near the disc periphery.

wt%

Ti

Mo

Darker bands Lighter bands

85.65 83.26

14.35 16.74

periphery, grains are not clearly distinguishable and lighter bands can be observed (Fig. 6c). The microstructure of the specimens with higher imposed strain (N > 1/4) could not be resolved by SEM due to the strong grain refinement. Refined microstructure of Ti-15Mo alloy prepared by HPT was shown by transmission electron microscopy in our previous study [38]. The initial stages of the microstructure refinement by HPT were investigated by the EBSD. Fig. 7. shows a high-resolution inverse pole figure (IPF) map from the centre part of the Ti-15Mo alloy after N =¼ HPT turn. High (θ > 15°) and low-angle grain boundaries (5° < θ < 15°) are highlighted in black and red color, respectively (θ denotes the misorientation angle). The microstructure consists of heavily deformed and twinned grains. The fragmentation of grains and formation of a UFG microstructure is visible especially in the right part of the image. Black areas in IPF map correspond to the points which could not be successfully indexed due to high deformation. In the Fig. 8, the IPF image from the middle part of the TIMETAL LCB alloy after N =¼ HPT rotation is shown. Regular black squares in the image are indents from the preceding microhardness measurement. Lattice rotation within grains and significant twinning can be observed. Twinning-induced grain refinement is clearly visible in the bottom right corner.

3.2.1. Ti-15Mo Ti-15Mo alloy in the β solution treated condition before HPT processing (not shown) consists of equiaxed grains with the average size of 20 – 50 µm. In the Fig. 5. the microstructure of the Ti-15Mo alloy sample after N =¼ HPT turns is shown. Comparatively large (~20 µm) heavily deformed grains were observed in the disc centre (Fig. 5a). In the “middle” region the microstructure is more deformed and inhomogeneous consisting of bigger grains surrounded by small grains (Fig. 5b). On the disc periphery, a refined grain structure is observed (Fig. 5c). Contrast in the SEM images is given predominantly by the channelling of electrons along differently oriented crystallographic planes. However, it was proven by energy dispersive X-ray (EDX) analysis that the observed bands in Fig. 5c are caused by segregation of Mo atoms. The average chemical composition of lighter and darker parts in Fig. 5 is summarized in Table 1. It is obvious that darker parts contain less Mo. The same segregation of Mo atoms is claimed in [37]. However, the mechanism of element partitioning is unknown. 3.2.2. TIMETAL LCB The cast and solution treated TIMETAL LCB alloy consists of very large grains with size up to 1 mm (not shown). The microstructure of the TIMETAL LCB alloy after N =¼ HPT turn is depicted in the Fig. 6. In the centre part, original large though deformed grains were observed (Fig. 6a). In the “middle” part, bigger grains are surrounded by small grains and the microstructure is clearly heavily twinned [32]. In the

3.3. Measurement of elastic constants of β-Ti alloys Elastic constants of Ti-15Mo and TIMETAL LCB alloys in the

Fig. 5. SEM images of Ti-15Mo alloy after N =¼ HPT turns (channelling contrast). a) centre. b) “middle”. c) periphery.

223

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Fig. 6. SEM images of TIMETAL LCB alloy after N =¼ HPT turns (channelling contrast). a) centre. b) “middle”. c) periphery. Table 2 Young's modulus and Poisson's ratio of Ti-15Mo and TIMETAL LCB alloys after different number of HPT turns.

Ti−15Mo ST Ti−15Mo N=1 Ti−15Mo N=5 TIMETAL LCB ST TIMETAL LCB N=1 TIMETAL LCB N=5

Youngʼs modulus E (GPa)

Poissonʼs ratio ν

Centre

Centre

91.0 98.1 96.6 80.1 91.2 89.7

Periphery

92.8 91.4 88.6 86.7

0.361 0.348 0.357 0.38 0.362 0.367

Periphery

0.359 0.367 0.367 0.377

solution treated (ST) condition and after N =1 and 5 HPT turns were measured. Table 2 contains results of the Young's modulus and the Poissonʼs ratio ν of alloy after different number of HPT rotations and in different regions of the disc. The standard deviations of the Young's modulus and the Poissonʼs ratio is ± 1 GPa and ± 0.01, respectively.

4. Discussion Fig. 7. Inverse pole figure map from the centre part of the Ti-15Mo alloy after N =¼ HPT turn; misorientation measured along the highlighted black line. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

4.1. Microhardness The microhardness of the Ti-15Mo alloy increases from HV ≈310 (solution treated coarse-grained material) to HV ≈450 (heavily deformed material after HPT). Maximum strength achieved in a ductile α+β aged condition is 1150 MPa [39], which corresponds approximately to HV ≈390. HPT deformation of the β ST condition therefore results in higher hardness than in the α+β aged condition. Contrary to the precipitation hardening, SPD increases the strength while retaining the β phase with lower Young's modulus. Achieving high strength and low modulus simultaneously is crucial for potential application of the material in biomedicine as an implant material. Vickers microhardness of the β ST condition of the TIMETAL LCB alloy HV ≈340 is slightly higher than that of Ti-15Mo alloy due to solid solution strengthening by Fe and Al as reported in [40,41]. The microhardness achieved by HPT deformation exceeds the hardness of the aged two-phase α+β condition [42]. Three types of strength evolution with increasing strain during HPT were proposed for different metals and alloys: strain hardening without recovery, strain hardening with recovery and strain softening [43]. In the present study, strain hardening without recovery is dominant since recovery processes in Ti alloys are not activated at room temperature. The microhardness increase due to imposed equivalent strain can be modelled by empirical work hardening models such as Hollomon model [44] or Voce model [45]. The Von Mises equivalent strain by HPT was calculated according to the Eq. (1) and the evolution of microhardness was depicted for the Ti-15Mo alloy (Fig. 9) and the TIMETAL LCB alloy (Fig. 10). All data points were used for the calculation, i.e. both from the measurement on the disc surface and

Fig. 8. Inverse pole figure map of TIMETAL LCB alloy after ¼ HPT turn (middle part). (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

224

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Table 3 Fitting parameters for Ti-15Mo, TIMETAL LCB and Ti-6Al-4V alloys.

Ti−15Mo TIMETAL LCB Ti−6Al-4V[43]

HV0

HVmax

m

K

L

358 372 311

458 461 364

0.44 0.46 1.4×10−8

2.13 3.68 3.87

0.13 0.12 0.06

Consequently, none of the models resulted in a satisfactory fit. Therefore, the sum of the Hollomon-type and Voce-type model (hereinafter referred to as Hollomon-Voce model) was used for modelling of hardening of studied alloys (Eq.(4)): V V HV (ε ) = HVmax + HV0H + K · ε m − (HVmax − HV0V ) exp (−L·ε )

Resulting fits are shown in Figs. 9 and 10 (red line). The fits by the Hollomon model (blue line) and the Voce model (green line) are also displayed. It is clear, that the combined Hollomon-Voce curve matches the experimental data much better than the individual models. Fitting parameters observed from Hollomon-Voce model are listed in Table 3. Initial microhardness HV0 represents the sum of fitting parameters HV0H and HV0V . HVmax represents the sum of maximum microhardness of V m + K ·εmax . For comparison, anathe fitting function, i.e. HVmax = HVmax logous fitting procedure was employed for the microhardness data for the Ti-6Al-4 V alloy deformed by HPT reported in [43]. Note, that the HV0 values of 358 and 372 HV from the fit for the Ti15Mo and the TIMETAL LCB alloys, respectively, are significantly higher than the values measured for the initial coarse-grained material (310 and 340 HV for the studied alloys, respectively). The reason is that the increase of microhardness at low strains is very sharp and HV0 value is sensitive to fitting procedure. Additionally, neglecting the reduction of sample thickness in the calculation of the Von Mises strain may also affect the result of fitting at low strains. Negligible Hollomon hardenability exponent m for the Ti-6Al-4V alloy [43] means that the microhardness increase in this case can be completely modelled by Voce model. The strengthening capability of the studied β-Ti alloys is much higher than that of Ti-6Al-4V alloy and therefore SPD processing of solution treated β-Ti alloys is even more beneficial than in the case of pure Ti or α+β alloys. Very high microhardness of the investigated β-Ti alloys can be attributed to the increased dislocation density [38] and the grain fragmentation. Strengthening can be also caused by deformation-induced ω-phase [46,47]. However, the confirmation of this hypothesis needs further investigation which is beyond the scope of this paper.

Fig. 9. Evolution of the microhardness with equivalent strain in Ti-15Mo alloy. The red line denotes the fitted microhardness according to the Eq. (4), blue and green line according to the Eq. (2) and Eq. (3), respectively. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 10. Variation of the measured microhardness values with equivalent strain of TIMETAL LCB alloy. The red line denotes the fitted microhardness according to the Eq. (4), blue and green line according to the Eq. (2) and Eq. (3), respectively. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

cross-section. Data corresponding to the leaking off the material near the edge of the discs were neglected. Fitting with weighting was employed. According to [44], the hardness evolution with strain was first fitted by the Hollomon-type equation:

HV (ε ) = K ·ε m + HV0H ,

(4)

(2)

where ε is the Von Mises equivalent strain, HV(ε) is the measured Vickers microhardness of the material, m is the hardenability exponent, K is a material constant and HV0H is a fitting parameter which corresponds to the microhardness of the initial material. The Hollomon model (Eq. (2)) does not account for the microhardness saturation which occurs at high strains. Eq. (3) represents Vocetype equation, which assumes exponential saturation of microhardness V at a value HVmax : V V HV (ε ) = HVmax − (HVmax − HV0V ) exp (−L*ε ),

(3)

is a fitting parameter corresponding to the initial microwhere V is a fitting parameter corresponding to the hardness at ε=0, HVmax saturation value of the hardness and L can be regarded as the saturation rate. The microhardness of the studied alloys increases very fast with increasing Von Mises strain up to the strain ε ≈ 50 . However, the microhardness does not completely saturate even for very high strains.

HV0V

Fig. 11. Misorientation variations along the highlighted black line in Fig. 9. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

225

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

Fig. 12. The detail of the inverse pole figure from Fig. 7 (Ti-15Mo, N=1/4) showing the primary and secondary twins. a. Left figure: IPF image - the orientations of the matrix and twins are indicated. Right sketch: Graphical representation of the twinning system between the matrix and the twin. b. Left figure: The orientation of the primary and secondary twins are indicated. Right sketch: Graphical representation of the secondary twinning. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

4.2. Multiple twinning as the mechanism of grain refinement The activity of twinning system {1 1 2}〈1 1 1〉in Ti-15Mo alloy is clearly documented by EBSD mapping. Moreover, it can be shown that the primary twins undergo additional secondary twinning. It is concluded that multiple mechanical twinning significantly contributes to the microstructure refinement. EBSD observations prove the common mechanism of the grain refinement consisting of the combination of dislocation density accumulation, lattice rotation, formation of dislocation walls and subgrain boundaries. The lattice rotation within a grain was identified by misorientation variations along the black line shown in Fig. 7. The point-to-origin and point-to-point misorientations are shown in Fig. 11 by a black and red line, respectively. The misorientation within the grain increases up to 10° which suggests a high degree of imposed strain and a high density of dislocations stored in the material deformed by HPT. In the distances of approximately 42 µm and 55 µm from the selected origin, the point-to-point misorientation abruptly changes. It corresponds to the position of sub-grain boundaries, which are also clearly visible in Fig. 7 (marked by arrows). Detail analysis of the EBSD observations revealed an alternative mechanism of grain refinement, namely twinning-induced grain refinement. Fig. 12a shows the detail from the Fig. 7 along with a schematic illustration of the twin orientation. From the graphic representation it is clearly visible that twinning occurs in the system {1 1 2}〈1 1 1〉. Consequently, the misorientation between the parent matrix and the twin is approx. 55°. In the Fig. 12b the secondary twinning is shown and graphically

Fig. 13. Inverse pole figure map of twinned grain of Ti-15Mo alloy after N =¼ HPT turn. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

illustrated. It is unambiguously proven that the system of secondary twinning is again {1 1 2}〈1 1 1〉. Fig. 13 is a bigger excerpt from Fig. 7 showing that the multiple twinning in Ti-15Mo alloy contribute to grain refinement of material. The misorientation values prove that the twin shown in red is a primary twin formed from the matrix, whereas the twin shown in green is a secondary twin which was formed from the primary one. Twinning induced grain refinement was also reported in magnesium alloys (hcp structure) [48,49], stainless steel (fcc structure) [50],

226

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

copper (fcc strucutre) [51]. In commercially pure Ti (hcp structure) the twinning-induced grain refinement is well documented [52–54]. On the other hand, to our knowledge, grain refinement by multiple twinning has not yet been reported in metastable β-Ti alloys (bcc structure) deformed by SPD.

• •

4.3. Elastic constants of HPT deformed β-Ti alloys HPT deformation affects the elastic properties of the studied alloys. The Youngʼs modulus of Ti-15Mo alloy after N=1 HPT turn in the centre of the specimen (lowest deformation) is 98.1 GPa while the Youngʼs modulus of initial ST material was 91 GPa. The Youngʼs modulus decreases with further straining, i.e. with increasing number of HPT turns and increasing distance from the centre of the sample and the final Young's modulus (98.4) is comparable to that of ST material. Similar trend can be observed for TIMETAL LCB alloy: the Youngʼs modulus of HPT deformed material in the centre of the sample (91.2 GPa) is higher than the Youngʼs modulus of the ST condition (80.1 GPa). Increasing deformation leads to a decrease of Youngʼs modulus, but the final value for (N=5, periphery) remains higher than that of solution treated material. Alterations in Young's modulus might be caused by the deformation induced ω-phase observed in [37,55]. The Young's modulus of ω-phase reaches 200 GPa [56] and even a small fraction of ω-phase affects the elastic properties. According to [57], 10% ω-phase forms from the β-matrix after 1 HPT turn, while further deformation by 10 HPT turns leads to a reverse phase transformation. In the case of TIMETAL LCB, the recovery of the Young's modulus is not complete, which suggest that the deformation induced ω-phase is not fully reverted. A decrease of Young's modulus with increasing deformation might be further caused by increasing density of free-volume defects such as non-equilibrium grain boundaries and the density dislocations (see [57] and other references therein). The first HPT turn induces also a measurable decrease of the Poissonʼs ration, especially in the central part of samples. This can be again explained by formation of a small volume fraction of the ω-phase, as the isotropic Poissonʼs ratio for the ω-phase is apparently low, ν=0.27 [56]. With increasing deformation, the value of ν fully recovers. The experimental results from the measurement of the Youngʼs modulus of Ti-15Mo alloy are comparable with the results indicated in [58]. The increase of Young's modulus due to the formation of ω-phase is adverse for potential application in biomedicine, where low Young's modulus is required. It might be beneficial to use alloys less prone to ωphase formation, for instance Ti-Nb based alloys if low Young's modulus is the major concern. However, the Young's modulus of ultra-fine grained Ti-15Mo and TIMETAL LCB alloys with increased hardness is still significantly lower than the Young's modulus of Ti-6Al4V alloy [6].

Acknowledgement This work was financially supported by Czech Science Foundation under the project GB 14-36566G. K.V. acknowledges the student project of Grant Agency of Charles University GAUK No. 1106216 and the project SVV-2016–260213. References [1] G. Lütjering, J.C. Williams, Titanium, Springer-Verlag, Berlin, Heidelberg, 2003. [2] M. Niinomi, Mechanical properties of biomedical titanium alloys, Mater. Sci. Eng. A 243 (1998) 231–236. http://dx.doi.org/10.1016/S0921-5093(97)00806-X. [3] R.R. Boyer, An overview on the use of titanium in the aerospace industry, Mater. Sci. Eng. A 213 (1996) 103–114. http://dx.doi.org/10.1016/0921-5093(96)102331. [4] D. Kuroda, M. Niinomi, M. Morinaga, Y. Kato, T. Yashiro, Design and mechanical properties of new β type titanium alloys for implant materials, Mater. Sci. Eng. A 243 (1998) 244–249. http://dx.doi.org/10.1016/S0921-5093(97)00808-3. [5] D.F. Williams, Titanium for Medical Applications, Springer-Verlag, Berlin, Heidelberg, 2001. [6] M. Geetha, A.K. Singh, R. Asokamani, A.K. Gogia, Ti based biomaterials, the ultimate choice for orthopaedic implants – a review, Prog. Mater. Sci. 54 (2009) 397–425. http://dx.doi.org/10.1016/j.pmatsci.2008.06.004. [7] E.S.N. Lopes, R.J. Contieri, S.T. Button, R. Caram, Femoral hip stem prosthesis made of graded elastic modulus metastable β Ti alloy, Mater. Des. 69 (2015) 30–36. http://dx.doi.org/10.1016/j.matdes.2014.11.040. [8] M. Niinomi, Recent research and development in titanium alloys for biomedical applications and healthcare goods, Sci. Technol. Adv. Mat. 4 (2003) 445–454. http://dx.doi.org/10.1016/j.stam.2003.09.002. [9] C. Leyens, M. Peters, Titanium and Titanium Alloys, WILEY-VCH Verlag, Weinheim, 2003. [10] J.C. Fanning, Properties of TIMETAL 555 (Ti-5Al-5Mo-5V-3Cr-0.6Fe), J. Mater. Eng. Perform. 14 (2005) 788–791. http://dx.doi.org/10.1361/105994905×75628. [11] A.P. Zhilyaev, T.G. Langdon, Using high-pressure torsion for metal processing: fundamentals and applications, Prog. Mater. Sci. 53 (2008) 893–979. http:// dx.doi.org/10.1016/j.pmatsci.2008.03.002. [12] R.Z. Valiev, et al., Nanostructured SPD processed titanium for medical implants, Mater. Sci. Forum 584- 586 (2008) 49–54. [13] A.A. Popov, et al., Structural and mechanical properties of nanocrystalline titanium processed by severe plastic deformation, Scr. Mater. 37 (1997) 1089–1094. http:// dx.doi.org/10.1016/S1359-6462(97)00210-8. [14] R.Z. Valiev, Structure and mechanical properties of ultrafine-grained metals, Mater. Sci. Eng. A 234–236 (1997) 59–66. http://dx.doi.org/10.1016/S0921-5093(97) 00183-4. [15] I.P. Semenova, G.I. Raab, L.R. Saitova, R.Z. Valiev, The effect of equal-channel angular pressing on the structure and mechanical behaviour of Ti-6Al-4V alloy, Mater. Sci. Eng. A 387–389 (2004) 805–808. http://dx.doi.org/10.1016/ j.msea.2004.02.093. [16] M. Janeček, et al., Mechanical properties and dislocation structure evolution in Ti6Al7Nb alloy processed by high pressure torsion, Metall. Mater. Trans. A 45A (2014) 7–15. http://dx.doi.org/10.1007/s11661-013-1763-2. [17] L.R. Saitova, H.W. Hoppel, M. Göken, I.P. Semenova, R.Z. Valiev, Cyclic deformation behaviour and fatigue lives of ultrafine-grained Ti-6Al-4V ELI alloy for biomedical use, Int. J. Fatigue 31 (2009) 322–331. http://dx.doi.org/10.1016/ j.ijfatigue.2008.08.007. [18] V. Polyakova, I.P. Semenova, R.Z. Valiev, Influence of annealing on the structure and mechanical properties of ultrafine-grained alloy Ti-6Al-7Nb, processed by severe plastic deformation, Mater. Sci. Forum 667–669 (2010) 943–948. [19] Z.B. Zhang, Y.L. Hao, S.J. Li, R. Yang, Fatigue behaviour of ultrafine-grained Ti24Nb-4Zr-8Sn multifunctional biomedical titanium alloy, Mater. Sci. Eng. A 577 (2013) 225–233. http://dx.doi.org/10.1016/j.msea.2013.04.051. [20] D. Kent, G. Wang, Z. Yu, X. Ma, M. Dargusch, Strength enhancement of a biomedical titanium alloy through modified accumulative roll bonding technique, J. Mech. Behav. Biomed. Mater. 4 (2011) 405–416. http://dx.doi.org/10.1016/ j.jmbbm.2010.11.013. [21] T. Li, D. Kent, G. Sha, M. Dargusch, J.M. Cairney, Precipitation of the α-phase in an

5. Conclusions In the present work, the microstructure and the microhardness evolution in ultra-fine grained metastable β-titanium alloys prepared by HPT were investigated. The most important results of this investigation can be summarized as follows:

• • •

higher Mo content were identified in severely deformed Ti-15Mo alloy. Twinning system {1 1 2} < 111 > is active during HPT deformation. Multiple twinning contributes to the grain refinement in bcc β-Ti alloys. The Young's modulus’ variations with the strain due in the investigated β-Ti alloys were nonmonotonous. Young's modulus first increases after the initial stage of the HPT deformation, while it decreases with further straining. This can be explained by the enhanced stress-induced ω-phase formation in the initial stages of HPT straining and is further supported by the simultaneous decrease of the Poissonʼs ratio during the first HPT turn.

Microhardness of Ti-15Mo and TIMETAL LCB alloys increases with increasing von Mises equivalent strain. The highest microhardness of β solution treated material deformed by HPT exceeds significantly the microhardness of commonly used aged two phase α+β condition of the same alloys. The capability of deformation hardening by HPT in both studied βTi alloys is much higher than in common α+β Ti-6Al-4V alloy. Scanning electron microscopy demonstrated the increasing grain fragmentation with the increasing equivalent strain. Bands with 227

Materials Science & Engineering A 682 (2017) 220–228

K. Václavová et al.

[22]

[23]

[24]

[25]

[26] [27] [28]

[29] [30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40] J. Stráský, M. Janeček, P. Harcuba, M. Landa, Plastic deformation and elastic properties of Ti-Nb-Zr-Ta(-Fe-Si) biomedical alloys, Adv. Mater. Res. 922 (2014) 734–739. http://dx.doi.org/10.4028/www.scientific.net/AMR.922.734. [41] E.W. Collings, Materials Properties Handbook: Titanium Alloys, ASM International, 1993. [42] J. Šmilauerová, Phase Transformations and Microstructure Changes in TIMET LCB Alloy (Master thesis), Charles University in Prague, 2012. [43] Y.C. Wang, T.G. Langdon, Effect of heat treatment on microstructure and microhardness evolution in a Ti-6Al-4V alloy processed by high pressure torsion, J. Mater. Sci. 48 (2013) 4646–4652. http://dx.doi.org/10.1007/s10853-012-7071-1. [44] J.H. Hollomon, Trans. Metall. Soc. AIME 162 (1945) 268. [45] E. Voce, The relationship between stress and strain for homogeneous deformations, J. Inst. Met. 74 (1948) 537–562. [46] F. Sun, et al., Investigation of early stage deformation mechanisms in a metastable β titanium alloy showing combined twinning-induced plasticity and transformation-induced plasticity effects, Acta Mater. 61 (2013) 6406–6417. [47] S. Gatina, I. Semenova, M. Janeček, J. Strasky, Effect of high pressure torsion on the aging kinetics of β-titanium Ti-15Mo alloy, IOP Conf. Ser. Mater. Sci. Eng. 63 (2014). [48] Y. Li, et al., Grain refinement due to complex twin formation in rapid hot forging of magnesium alloy, Scr. Mater. 68 (2013) 171–174. http://dx.doi.org/10.1016/ j.scriptamat.2012.10.007. [49] H. Miura, M. Ito, X. Yang, J.J. Jonas, Mechanisms of grain refinement in Mg–6Al– 1Zn alloy during hot deformation, Mater. Sci. Eng. A. 538 (2012) 63–68. http:// dx.doi.org/10.1016/j.msea.2012.01.014. [50] H.W. Zhang, Z.K. Hei, G. Liu, J. Lu, K. Lu, Formation of nanostructured surface layer on AISI 304 stainless steel by means of surface mechanical attrition treatment, Acta Mater. 51 (2003) 1871–1881. http://dx.doi.org/10.1016/S13596454(02)00594-3. [51] K. Wang, N.R. Tao, G. Liu, J. Lu, K. Lu, Plastic strain-induced grain refinement at the nanometer scale in copper, Acta Mater. 54 (2006) 5281–5291. http:// dx.doi.org/10.1016/j.actamat.2006.07.013. [52] S.V. Zherebtsov, et al., Evolution of grain and subgrain structure during cold rolling of commercial-purity titanium, Mater. Sci. Eng. A. 528 (2011) 3474–3479. http:// dx.doi.org/10.1016/j.msea.2011.01.039. [53] D.H. Hong, S.K. Hwang, Microstructural refinement of CP-Ti by cryogenic channeldie compression involving mechanical twinning, Mater. Sci. Eng. A. 555 (2012) 106–116. http://dx.doi.org/10.1016/j.msea.2012.06.040. [54] C.S. Meredith, A.S. Khan, The microstructural evolution and thermo-mechanical behaviour of UFG Ti processed via equal channel angular pressing, J. Mater. Process. Technol. 219 (2015) 257–270. http://dx.doi.org/10.1016/j.jmatprotec.2014.12.024. [55] B. Koch, B. Skrotzki, Strain controlled fatigue testing of the metastable β-titanium alloy Ti–6.8Mo–4.5Fe–1.5Al (Timetal LCB), Mater. Sci. Eng. A 528 (2011) 5999–6005. http://dx.doi.org/10.1016/j.msea.2011.04.031. [56] M. Tane, Y. Okuda, Y. Todaka, H. Ogi, A. Nagakubo, Elastic properties of singlecrystalline ω phase in titanium, Acta Mater. 61 (2013) 7543–7554. [57] A.B. Lebedev, Y.A. Burenkov, A.E. Romanov, V.I. Kopylov, V.P. Filonenko, V.G. Gyaznov, Softening of the in submicrocrystalline copper, Mater. Sci. Eng. A 203 (1995) 165–170. [58] S. Gatina, I. Semenova, J. Leuthodl, R. Valiev, Nanostructuring and phase transformations in the β-alloy Ti-15Mo during high-pressure torsion, Adv. Eng. Mater. 17 (2015) 1742–1747.

ultrafine grained beta-titanium alloy processed by severe plastic deformation, Mater. Sci. Eng. A 605 (2014) 144–150. http://dx.doi.org/10.1016/ j.msea.2014.03.044. D. Kent, W.L. Xiao, G. Wang, Z. Yu, M.S. Dargusch, Thermal stability of an ultrafine grain β-Ti alloy, Mater. Sci. Eng. A 556 (2012) 582–587. http://dx.doi.org/ 10.1016/j.msea.2012.07.030. M. Geetha, A.K. Singh, R. Asokamani, A.K. Gogia, Ti based biomaterials, the ultimate choice for orthopaedic implants – a review, Prog. Mater. Sci. 54 (2009) 397–425. Kelvin Y. Xie, et al., Nanocrystalline β-Ti alloy with high hardness, low Young's modulus and excellent in vitro biocompatibility for biomedical applications, Mater. Sci. Eng. C. 33 (2013) 3530–3536. I. Weiss, S.L. Semiatin, Thermomechanical processing of beta titanium alloys – an overview, Mater. Sci. Eng. A. 243 (1998) 46–65. http://dx.doi.org/10.1016/S09215093(97)00783-1. G. Gottstein, Physical Foundations of Materials Science, Springer-Verlag, 2004. H. Paxton, Experimental verification of the twin system in alpha-iron, Acta Metall. 1 (1953) 141–143. http://dx.doi.org/10.1016/0001-6160(53)90052-2. J.W. Christian, S. Mahajan, Deformation twinning and its effect on crack extension, Prog. Mater. Sci. 39 (1995) 1–157. http://dx.doi.org/10.1016/0079-6425(94) 00007-7. Y. Wang, et al., Deformation twinning during nanoindentation of nanocrystalline Ta, Appl. Phys. Lett. 86 (2005) 101915. http://dx.doi.org/10.1063/1.1883335. Y. Yang, S. Wu, G. Li, Y. Li, Y. Lu, K. Yang, P. Ge, Evolution of deformation mechanisms of Ti–22.4Nb–0.73Ta–2Zr–1.34O alloy during straining, Acta Mater. 58 (2010) 2778–2787. http://dx.doi.org/10.1016/j.actamat.2010.01.015. X.H. Min, et al., {3 3 2}〈1 1 3〉Twinning system selection on a β-type Ti-15Mo5Zr polycrystalline alloy, Mater. Sci. Eng. A. 579 (2013) 164–169. http:// dx.doi.org/10.1016/j.msea.2013.04.119. X.H. Min, K. Tsuzaki, S. Emura, K. Tsuchiya, Heterogeneous twin formation and its effect on tensile properties in Ti-Mo based β titanium alloys, Mater. Sci. Eng. A. 554 (2012) 53–60. http://dx.doi.org/10.1016/j.msea.2012.06.009. R.Z. Valiev, A.V. Korznikov, R.R. Mulyukov, Structure and properties of ultrafinegrained materials produced by severe plastic deformation, Mater. Sci. Eng. A168 (1993) 141–148. R.Z. Valiev, et al., Structure and deformation behaviour of Armco iron subjected to severe plastic deformation, Acta Mater. 44 (1996) 4705–4712. http://dx.doi.org/ 10.1016/S1359-6454(96)00156-5. A. Migliori, J.L. Sarrao, W.M. Visscher, T.M. Bell, M. Lei, Z. Fisk, R.G. Leisure, Resonant ultrasound spectroscopic techniques for measurement of the elastic moduli of solids, Physica B Phys. Condens Matter 183 (1993) 1–24. P. Sedlák, H. Seiner, J. Zídek, M. Janovská, M. Landa, Determination of all 31 Independent Elastic Coefficients of Generally Anisotropic Solids by Resonant Ultrasound, Spectrosc.: Benchmark Ex. Exp. Mech. 54 (2014) 1073–1085. http:// dx.doi.org/10.1007/s11340-014-9862-6. B. Jiang, K. Tsuchiya, S. Emura, X. Min, Effect of high-pressure torsion process on precipitation behaviour of α phase in β-type Ti–15Mo alloy, Mater. Trans. 55 (2014) 877–884. M. Janeček, et al., Microstructure evolution in solution treated Ti15Mo alloy processed by high pressure torsion, Mater. Charact. 98 (2014) 233–240. http:// dx.doi.org/10.1016/j.matchar.2014.10.024. ATI 15Mo Titanium Alloy Technical Sheet, 〈https://www.atimetals.com/ Documents/ati_15Mo_Titanium_Alloy_en_v4%20final.pdf〉.

228