- Email: [email protected]
Strain rate sensitivity and deformation activation volume of coarse-grained and ultrafine-grained TiNi alloys
Available online at www.sciencedirect.com
ScienceDirect Scripta Materialia 102 (2015) 99–102 www.elsevier.com/locate/scriptamat
Strain rate sensitivity and deformation activation volume of coarse-grained and ultrafine-grained TiNi alloys ⇑
D.V. Gunderov,a,b,d, G. Maksutova,a A. Churakova,a,b A. Lukyanov,a A. Kreitcberg,c G.I. Raab,a,d I. Sabirove and S. Prokoshkinc a
Ufa State Aviation Technical University, Ufa, Russia Institute of Molecule and Crystal Physics RAS, Ufa, Russia c National University of Science and Technology MISIS, Moscow, Russia d Kazan Federal University, Kazan, Russia e IMDEA Materials Institute, Madrid, Spain b
Received 31 August 2014; revised 27 January 2015; accepted 14 February 2015 Available online 3 March 2015
Two TiNi alloys, Ti49.4Ni50.6 and Ti50Ni50, are subjected to equal-channel angular pressing (ECAP) resulting in the formation of a homogeneous ultra-fine grained microstructure. Tensile tests and strain rate jump tests are carried out in the temperature range of 25–400 °C to measure mechanical properties and strain rate sensitivity of both alloys before and after ECAP processing. Effect of grain size on mechanical behavior, strain rate sensitivity and mechanisms operating during plastic deformation of both alloys in the given temperature range is discussed. Ó 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: TiNi alloys; Severe plastic deformation; Ultra-fine grained material; Strain rate sensitivity; Activation volume
Titanium-nickel (TiNi) based alloys refer to the class of functional materials with the shape-memory effect (SME), stipulated by thermoelastic martensitic phase transformations [1,2]. These alloys are widely used as construction and functional materials in engineering and medicine [2,3]. It has been demonstrated over the last decade that grain refinement down to ultrafine- or nano-scale is a promising direction to enhance their mechanical and functional properties [4–11]. However, research on mechanical behavior and mechanisms operating during plastic deformation of the UFG TiNi alloys is limited [6,12]. Analysis of the mechanisms responsible for the material plastic flow at different temperatures is possible based on the data on strain rate sensitivity m and correspondingly calculated activation volume V [13]. Based on the analysis of activation volume, one may suggest a hypothesis about controlling deformation mechanisms in the UFG TiNi alloys. Moreover, manipulation with strain rate sensitivity in UFG materials is considered as one of the strategies to improve their ductility, since enhanced strain rate sensitivity can significantly improve their tensile ductility [14]. However, up to date, there are no available literature data
⇑ Corresponding
author at: Institute of Physics of Advanced Materials, Ufa State Aviation Technical University, 12 K. Marx str., Ufa 450000, Russia. Tel.: +7 347 2734449; fax: +7 347 2733422; e-mail: [email protected]
on the effect of grain size on strain rate sensitivity and activation volume in the UFG TiNi alloys. It is known that activation volume in materials with bccand hcp-lattices typically does not change with grain size, but in the case of fcc metals V decreases with decreasing grain size [13]. The intermetallic compound TiNi in the initial high-temperature austenite state has an atomic ordered structure B2 of CsCl type with parameters a = 0.301 nm, i.e. a bcc lattice. During cooling or under the applied stresses the B2-phase can transform into B19’ phase with a monocline distorted orthorhombic elementary lattice with parameters a = 0.289 nm, b = 0.412 nm, c = 0.462 nm, b = 97° [1,2]. Thus, below the temperatures of suppression of martensitic transformation (Md), TiNi alloys are deformed in martensitic state (B19’ phase); whereas at temperatures over Md, TiNi alloys are deformed in austenitic state (B2 phase). The main objective of this work is to study the effect of grain size and temperature (i.e. phase) on strain rate sensitivity and activation volume of TiNi alloys. Two TiNi alloys were chosen as the materials for this investigation – Ti49.4Ni50.6 and Ti50Ni50. The Ti49.4Ni50.6 alloy (manufactured by Intrinsic Devices, USA) is overstoichiometric. The alloy belongs to the class of medical materials. At room temperature, it has the atomic-ordered B2 structure. The temperatures for martensite formation in the alloy are the following: Ms = 11 °C, Af = 42 °C. After quenching and further heating, the Ti49.4Ni50.6 alloy may
http://dx.doi.org/10.1016/j.scriptamat.2015.02.023 1359-6462/Ó 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
100
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
exhibit decomposition of solid solution with a gradual formation of second phase precipitates [2], such as Ti3Ni4, Ti2Ni3 and TiNi3. The Ti50Ni50 alloy (manufactured by Johnson Matthey, USA) is stoichiometric, non-aging. The temperatures for martensite formation in this alloy are the following: Ms = 73 °C, Af = 100 °C. At room temperature, the Ti50Ni50 alloy has the martensite B19’ structure. Bars with a diameter of 18 mm and a length of 100 mm of both alloys were solution treated at 800 °C for 2 h and quenched in water to generate a homogeneous coarsegrained microstructure. The bars of the Ti50Ni50 alloy were subjected to equal-channel angular pressing (ECAP) processing for 8 passes at 400 °C, whereas the Ti49.4Ni50.6 alloy was pressed at 450 °C for 8 passes [15,16]. The angle of channels’ intersection in the ECAP die was 110°. The microstructure of the samples was studied using transmission electron microscopy (TEM) on a JEM-2000 b microscope. Foils for TEM studies were prepared using the device for double-sided electropolishing TenuPol-5 according to the standard procedure. The electrolyte had the chemical composition 10%HClO4 + 90%CH3(CH2)3OH. Small flat tensile specimens with a gage area of 1 0.25 4 mm were used for the tests. Then, the specimens were ground and polished using standard metallographic techniques. Tensile specimens were deformed to failure with constant crosshead speed corresponding to initial strain rate e1 ¼ 1 103 s1 at test temperatures of 25 °C, 150 °C, 250 °C and 400 °C. Three samples were deformed at each condition and the results thus obtained were found to be reproducible. Prior to the start of a tensile test, the specimen was kept for 10 min at the preset test temperature. The test temperature was maintained within ±1 °C, as measured by a thermocouple clamped to the upper shoulder of the tensile specimen. To estimate the strain rate sensitivity (m), strain rate jump tests were performed at a strain rate reduced to e2 ¼ 1 104 s1 . The strain rate sensitivity index m was estimated using the relation [17] m ¼ lgðr2 =r1 Þ=lgðe2 =e1 Þ
ð1Þ
where r1 is the flow stress at the strain rate e1 (before a strain rate jump) and r2 is the flow stress at the strain rate e2 (after a strain rate jump). The activation volume V was estimated on the basis of the relation [13], p V ¼ 3kT =mr ð2Þ where k is Boltzmann’s constant, T the testing temperature and r the flow stress. All values of the activation volume reported in this article are given in a normalized form V/b3 (in units of b3, where b is the magnitude of the Burgers vector1). According to the optical metallographic examination using an OLYMPUS BX51 microscope, both Ti49.4Ni50.6 and Ti50Ni50 alloys after quenching have a homogeneous coarse-grained microstructure with an average grain size of 200 lm (Fig. 1). In both alloys, impurities in form of
1
According to the calculations on the lattice’s parameter, the Burgers ˚ . While being tested, TiNi can vector of the B2 phase is 2.61 A undergo the B2–B190 phase transformation (during deformation at temperatures lower than Md). During deformation at temperatures above Md, the B2–B190 phase transformation is being suppressed. For the sake of simplicity, the Burgers vector b of the B2 phase of ˚ was used to calculate V [b3] for all conditions. 2.61 A
200 µm
(a)
200 µm
(b)
Figure 1. Microstructure of the quenched alloys Ti49.4Ni50.6 (a) and Ti50Ni50 (b).
(a)
(b)
Figure 2. Microstructure after ECAP processing of: (a) Ti49.4Ni50.6 alloy, (b) Ti50Ni50 alloy.
Ti4Ni2OX spherical inclusions are observed. Since the Ti50Ni50 alloy is a martensitic one (Ms = 78 °C) at room temperature, martensite twins are also observed (Fig. 1b). The Ti49.4Ni50.6 alloy after ECAP processing is in the austenitic state and has a grain size of 0.3 lm (Fig. 2a). The Ti50Ni50 alloy after ECAP is in the martensitic state. From TEM images, the size of austenite grains after ECAP processing cannot be precisely measured due to austenite–martensite phase transformation during cooling after ECAP. Therefore, the size of austenite grains formed during ECAP processing of the Ti50Ni50 alloy was measured as the size of blocks of martensite within individual prior austenite grains. An approximate estimation of the austenite grain size after ECAP processing yields was 0.7 lm (Fig. 2b). A larger size of austenite grains in the ECAP processed Ti50Ni50 alloy as compared to the Ti49.4Ni50.6 alloy can be explained by the fact that the Ti49.4Ni50.6 alloy is an age-hardenable one and the fine particles, which precipitate at the temperature of ECAP processing, can suppress grain growth. Figure 3 demonstrates stress–strain curves of Ti49.4Ni50.6 alloys in CG and UFG states at temperatures of 25°, 150°, 250° and 400 °C. Stress–time plots from strain rate jump
Figure 3. Stress–strain curves from tensile testing of Ti50Ni50 alloy in the CG (a) and UFG (b) conditions.
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
Figure 4. Stress–time curves from strain rate jump testing of Ti49.4Ni50.6 alloy in the CG (a) and UFG (b) states.
testing experiments (where strain rate was changed from 103 s1 to 104 s1) are presented in Figure 4. The results of analysis of stress–strain curves and strain rate jump tests are listed in Table 1, where the average data on yield stress (YS), ultimate tensile strength (UTS), strain sensitivity m, and activation volume V are presented for the Ti50Ni50 and Ti49.4Ni50.6 alloys in the CG and UFG state. During tensile testing at room temperature, a pseudo yield area at rm stresses about 200–400 MPa (depending on the composition and state of the alloy) is observed (Fig. 3). This testifies either to a strain-induced B2–B190 phase transformation (in the Ti49.4Ni50.6 alloy), or to re-orientation of the martensite under applied stress (in the Ti50Ni50 alloy). No pseudo yield area is observed on tensile curves at temperatures above 150 °C. This testifies either to the B2–B190 phase transformation occurring simultaneously with the plastic deformation or to the suppression of the B2–B190 phase transformation. ECAP leads to a considerable increase of the ultimate tensile strength and yield stress of both alloys as compared with the initial quenched state. The UTS and YS of the age-hardenable Ti49.4Ni50.6 alloy in CG and UFG states are somewhat higher compared to those of the Ti50Ni50. Tensile testing of the alloys at room temperature in both CG and UFG states is characterized by a long stage of uniform deformation, which is related to the long-term hardening and twinning of the martensite phase [10,18,19]. The tensile testing at room temperature is finalized by a failure of samples without localization of plastic deformation and without formation of any neck. Strength and ductility of both alloys tend to decrease with increasing testing
101
temperature. Decrease of ductility with increasing temperature can be related to the suppression of the B2–B190 phase transformation, reduction of the uniform deformation and its faster localization. The strain rate sensitivity m increases with increasing testing temperature in both alloys in CG and UFG states. Depending on the composition and testing temperature, the m parameter in the UFG state is by a factor of 1.5–3 higher than the m parameter of the alloys in the CG state. The m parameter of the Ti50Ni50 alloy is also somewhat higher than m of the Ti49.4Ni50.6 alloy at all testing temperatures. The parameter m measured at RT has a considerably small value (0.002–0.007). This may be associated with the fact that twinning plays a significant role in the deformation of TiNi. It should be also noted that twinning has a negligible strain rate sensitivity, and a material that deforms primarily through twinning would be expected to have an abnormally low m compared to a material that deforms from dislocation mechanisms at any given temperature [21]. The activation volume V in the CG and UFG alloy Ti50Ni50 and the UFG alloy Ti49.4Ni50.6 grows with increasing temperatures to the maximum value (at T = 150°– 250 °C) and then drops at higher temperatures. In most of the cases (excluding the CG Ti49.4Ni50.6 alloy), a temperature increase up to 400 °C results in reduction of activation volume. In the CG Ti49.4Ni50.6 alloy, activation volume grows with increasing temperatures up to 400 °C. The parameter V is determined by a relationship between the values T, m and r (Eq. (2)). The strain rate sensitivity m increases with increasing testing temperature for all the alloys in all states. However, for the Ti50Ni50 alloy the parameter m grows with increasing temperature faster than for the Ti49.4Ni50.6 alloy. The dependence of the flow stress r on the testing temperature T is ambiguous and differs for alloys/state. All this accounts for the complexity of the dependence of the V-value on the testing temperature T for different alloys/states. The growth of the activation volume simultaneously with increasing testing temperature for T = 150°–250 °C could be rationalized based on the fact that tensile testing of both alloys at room temperature occurs in the martensitic state, whereas austenite–martensite phase transformation is suppressed at elevated temperatures.
Table 1. Results of mechanical testing of Ti50Ni50 and Ti49.4Ni50.6 alloys in CG and UFG states. d (%)
m
V, b3
823 494 471 318
59 36 35 17
0.004 0.004 0.01 0.03
169 419 461 119
870 584 598 703
1012 831 720 736
38 27 19 4
0.007 0.01 0.02 0.05
52 154 77 30
25 150 250 400
477 552 597 543
945 768 792 626
30 15 15 11
0.002 0.003 0.004 0.004
269 319 284 604
25 150 250 400
1160 906 948 1020
1256 1159 1094 1091
18 8 10 8
0.006 0.007 0.007 0.012
121 119 105 99
Alloy/state
Temperature (°C)
YS (MPa)
Ti50Ni50/CG
25 150 250 400
430 377 364 287
Ti50Ni50/ECAP
25 150 250 400
Ti49.4Ni50.6/CG
Ti49.4Ni50.6/ECAP
UTS (MPa)
102
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
Reduction of activation volume in most of the cases (excluding the CG Ti49.4Ni50.6 alloy) with a temperature increase up to 400 °C can be related to the enhanced contribution of diffusion and intergranular sliding at this temperature [20]. The different forms of the V(T) dependence for the CG and UFG Ti49.4Ni50.6 alloy with increasing temperature up to 400 °C could be explained by the fact that the kinetics of decomposition of solid solution and formation of precipitates in the UFG Ti-Ni age-hardenable alloys is significantly different from that in their CG counterparts [15], and these processes have a noticeable effect on deformation. However, to account for the observed V(T) dependence for CG and UFG TiNi with different compositions, additional studies are definitely required. It is known that activation volume within a range of 1– 10 b3 points at diffusion and grain boundary sliding as the controlling deformation mechanisms [13,17]. Annihilation of dislocations at grain boundaries dominates plastic deformation if the V value is within a range of 10–100 b3 [17]. In case the V value is within a range of 100–1000 b3, the overcoming of the forest junctions by gliding dislocations is considered as the controlling deformation mechanism [13,17]. Therefore, according to the V value, within the studied range of temperatures in the CG state of TiNi (V 100–400 b3), thermally activated overcoming of forest junctions by moving dislocations controls thermally activated slip. In case of the UFG TiNi alloys (V 20– 120 b3), plastic deformation is mainly controlled by annihilation of dislocations along grain boundaries. In summary: ECAP leads to a considerable increase of the UTS and YS of both TiNi alloys as compared with the initial state. The UTS and YS of the age-hardenable Ti49.4Ni50.6 alloy in CG and UFG states were somewhat higher compared to those of the Ti50Ni50. Tensile testing of the alloys at room temperature is characterized by a long stage of uniform deformation, a failure of samples took place without localization of plastic deformation. With increasing testing temperature ductility decreased, that related to the suppression of the B2–B190 phase transformation and faster deformation localization. The UFG TiNi alloys have higher strain rate sensitivity and lower activation volume compared to their coarsegrained counterparts in the temperature range of 20– 400 °C. Strain rate sensitivity tends to increase with increasing temperature in CG and UFG TiNi alloys, whereas the effect of temperature on activation volume is ambiguous. This is accounted for by the complexity of phase transformations in the material at increasing testing temperature T (martensite–austenite transformation, solid solution decomposition, aging), which has a complex effect on flow stress.
D.V. Gunderov acknowledges gratefully funding through the Russian Government Program of Competitive Growth of Kazan Federal University. G.I. Raab acknowledges gratefully financial support by the Russian Ministry of Education and Science of Russia through project No. 1.729.2014R.
[1] V. Brailovski, S. Prokoshkin, P. Terriault, F. Trochu (Eds.), Shape Memory Alloys: Fundamentals, Modeling and Applications, Ecole de Technologie Superieure (ETS), Universite du Quebec, Montreal, Canada, 2003, 851p. [2] K. Otsuka, X. Ren, Prog. Mater. Sci. 50 (2005) 511. [3] L. Mishnaevsky Jr., E. Levashov, R.Z. Valiev, J. Segurado, I. Sabirov, N. Enikeev, S. Prokoshkin, A.V. Solov’yov, A. Korotitskiy, E. Gutmanas, I. Gotman, E. Rabkin, S. Psakh’e, L. Dluhos, M. Seefeldt, A. Smolin, Mater. Sci. Eng. R 81 (2014) 1. [4] T. Waitz, V. Kazykhanov, H.P. Karnthaler, Acta Mater. 52 (2004) 137. [5] B. Kockar, I. Karaman, J.I. Kim, Y.I. Chumlyakov, J. Sharp, C.J. Yu, Acta Mater. 56 (2008) 3630. [6] M.L. Young, M.F.X. Wagner, J. Frenzel, W.W. Schmahl, G. Eggeler, Acta Mater. 58 (2010) 2344. [7] V.G. Pushin, R.Z. Valiev, Y.T. Zhu, D.V. Gunderov, T.E. Kuntsevich, N.I. Kourov, A.N. Uksusnicov, L.I. Yurchenko, Mater. Trans. 47 (2006) 694. [8] A.V. Sergueeva, C. Song, R.Z. Valiev, A.K. Mukherjee, Mater. Sci. Eng. A 339 (2003) 159. [9] V.G. Pushin, V.V. Stolyarov, R.Z. Valiev, T.C. Lowe, Y.T. Zhu, Mater. Sci. Eng. A 410–411 (2005) 386. [10] R.Z. Valiev, D.V. Gunderov, A.V. Lukyanov, V.G. Pushin, J. Mater. Sci. 47 (2012) 7848. [11] I.Yu. Khmelevskaya, S.D. Prokoshkin, I.B. Trubitsyna, M.N. Belousov, S.V. Dobatkin, E.V. Tatyanin, A.V. Korotitskiy, V. Brailovski, V.V. Stolyarov, E.A. Prokofiev, Mater. Sci. Eng. A 481–482 (2008) 119. [12] I. Karaman, A.V. Kulkarni, Z.P. Luo, Philos. Mag. 85 (2005) 1729. [13] P. Rodriguez, Metall. Mater. Trans. A. 35 (2004) 2697. [14] Y. Zhao, Y.T. Zhu, E.J. Lavernia, Adv. Eng. Mater. 12 (2010) 769. [15] E.A. Prokofiev, J.A. Burow, E.J. Payton, R. Zarnetta, J. Frenzel, D.V. Gunderov, R.Z. Valiev, G. Eggeler, Adv. Mater. Eng. 12 (2010) 747. [16] D. Gunderov, A. Lukyanov, E. Prokofiev, A. Churakova, V. Pushin, S. Prokoshkin, V. Stolyarov, R. Valiev, Mater. Sci. Forum 738–739 (2013) 486. [17] D. Caillard, J.L. MartinThermally Activated Mechanisms in Crystal Plasticity, Pergamon Materials Series, vol. 8, Elsevier, Oxford, United Kingdom, 2003. [18] J.X. Zhang, M. Sato, A. Ishida, Acta Mater. 54 (2006) 1185. [19] J.X. Zhang, M. Sato, A. Ishida, Acta Mater. 49 (2001) 3001. [20] M. Kawasaki, T.G. Langdon, J. Mater. Sci. 42 (2007) 1782. [21] C.H.J. Davies, Mater. Sci. Forum 539–543 (2007) 1723.
ScienceDirect Scripta Materialia 102 (2015) 99–102 www.elsevier.com/locate/scriptamat
Strain rate sensitivity and deformation activation volume of coarse-grained and ultrafine-grained TiNi alloys ⇑
D.V. Gunderov,a,b,d, G. Maksutova,a A. Churakova,a,b A. Lukyanov,a A. Kreitcberg,c G.I. Raab,a,d I. Sabirove and S. Prokoshkinc a
Ufa State Aviation Technical University, Ufa, Russia Institute of Molecule and Crystal Physics RAS, Ufa, Russia c National University of Science and Technology MISIS, Moscow, Russia d Kazan Federal University, Kazan, Russia e IMDEA Materials Institute, Madrid, Spain b
Received 31 August 2014; revised 27 January 2015; accepted 14 February 2015 Available online 3 March 2015
Two TiNi alloys, Ti49.4Ni50.6 and Ti50Ni50, are subjected to equal-channel angular pressing (ECAP) resulting in the formation of a homogeneous ultra-fine grained microstructure. Tensile tests and strain rate jump tests are carried out in the temperature range of 25–400 °C to measure mechanical properties and strain rate sensitivity of both alloys before and after ECAP processing. Effect of grain size on mechanical behavior, strain rate sensitivity and mechanisms operating during plastic deformation of both alloys in the given temperature range is discussed. Ó 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: TiNi alloys; Severe plastic deformation; Ultra-fine grained material; Strain rate sensitivity; Activation volume
Titanium-nickel (TiNi) based alloys refer to the class of functional materials with the shape-memory effect (SME), stipulated by thermoelastic martensitic phase transformations [1,2]. These alloys are widely used as construction and functional materials in engineering and medicine [2,3]. It has been demonstrated over the last decade that grain refinement down to ultrafine- or nano-scale is a promising direction to enhance their mechanical and functional properties [4–11]. However, research on mechanical behavior and mechanisms operating during plastic deformation of the UFG TiNi alloys is limited [6,12]. Analysis of the mechanisms responsible for the material plastic flow at different temperatures is possible based on the data on strain rate sensitivity m and correspondingly calculated activation volume V [13]. Based on the analysis of activation volume, one may suggest a hypothesis about controlling deformation mechanisms in the UFG TiNi alloys. Moreover, manipulation with strain rate sensitivity in UFG materials is considered as one of the strategies to improve their ductility, since enhanced strain rate sensitivity can significantly improve their tensile ductility [14]. However, up to date, there are no available literature data
⇑ Corresponding
author at: Institute of Physics of Advanced Materials, Ufa State Aviation Technical University, 12 K. Marx str., Ufa 450000, Russia. Tel.: +7 347 2734449; fax: +7 347 2733422; e-mail: [email protected]
on the effect of grain size on strain rate sensitivity and activation volume in the UFG TiNi alloys. It is known that activation volume in materials with bccand hcp-lattices typically does not change with grain size, but in the case of fcc metals V decreases with decreasing grain size [13]. The intermetallic compound TiNi in the initial high-temperature austenite state has an atomic ordered structure B2 of CsCl type with parameters a = 0.301 nm, i.e. a bcc lattice. During cooling or under the applied stresses the B2-phase can transform into B19’ phase with a monocline distorted orthorhombic elementary lattice with parameters a = 0.289 nm, b = 0.412 nm, c = 0.462 nm, b = 97° [1,2]. Thus, below the temperatures of suppression of martensitic transformation (Md), TiNi alloys are deformed in martensitic state (B19’ phase); whereas at temperatures over Md, TiNi alloys are deformed in austenitic state (B2 phase). The main objective of this work is to study the effect of grain size and temperature (i.e. phase) on strain rate sensitivity and activation volume of TiNi alloys. Two TiNi alloys were chosen as the materials for this investigation – Ti49.4Ni50.6 and Ti50Ni50. The Ti49.4Ni50.6 alloy (manufactured by Intrinsic Devices, USA) is overstoichiometric. The alloy belongs to the class of medical materials. At room temperature, it has the atomic-ordered B2 structure. The temperatures for martensite formation in the alloy are the following: Ms = 11 °C, Af = 42 °C. After quenching and further heating, the Ti49.4Ni50.6 alloy may
http://dx.doi.org/10.1016/j.scriptamat.2015.02.023 1359-6462/Ó 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
100
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
exhibit decomposition of solid solution with a gradual formation of second phase precipitates [2], such as Ti3Ni4, Ti2Ni3 and TiNi3. The Ti50Ni50 alloy (manufactured by Johnson Matthey, USA) is stoichiometric, non-aging. The temperatures for martensite formation in this alloy are the following: Ms = 73 °C, Af = 100 °C. At room temperature, the Ti50Ni50 alloy has the martensite B19’ structure. Bars with a diameter of 18 mm and a length of 100 mm of both alloys were solution treated at 800 °C for 2 h and quenched in water to generate a homogeneous coarsegrained microstructure. The bars of the Ti50Ni50 alloy were subjected to equal-channel angular pressing (ECAP) processing for 8 passes at 400 °C, whereas the Ti49.4Ni50.6 alloy was pressed at 450 °C for 8 passes [15,16]. The angle of channels’ intersection in the ECAP die was 110°. The microstructure of the samples was studied using transmission electron microscopy (TEM) on a JEM-2000 b microscope. Foils for TEM studies were prepared using the device for double-sided electropolishing TenuPol-5 according to the standard procedure. The electrolyte had the chemical composition 10%HClO4 + 90%CH3(CH2)3OH. Small flat tensile specimens with a gage area of 1 0.25 4 mm were used for the tests. Then, the specimens were ground and polished using standard metallographic techniques. Tensile specimens were deformed to failure with constant crosshead speed corresponding to initial strain rate e1 ¼ 1 103 s1 at test temperatures of 25 °C, 150 °C, 250 °C and 400 °C. Three samples were deformed at each condition and the results thus obtained were found to be reproducible. Prior to the start of a tensile test, the specimen was kept for 10 min at the preset test temperature. The test temperature was maintained within ±1 °C, as measured by a thermocouple clamped to the upper shoulder of the tensile specimen. To estimate the strain rate sensitivity (m), strain rate jump tests were performed at a strain rate reduced to e2 ¼ 1 104 s1 . The strain rate sensitivity index m was estimated using the relation [17] m ¼ lgðr2 =r1 Þ=lgðe2 =e1 Þ
ð1Þ
where r1 is the flow stress at the strain rate e1 (before a strain rate jump) and r2 is the flow stress at the strain rate e2 (after a strain rate jump). The activation volume V was estimated on the basis of the relation [13], p V ¼ 3kT =mr ð2Þ where k is Boltzmann’s constant, T the testing temperature and r the flow stress. All values of the activation volume reported in this article are given in a normalized form V/b3 (in units of b3, where b is the magnitude of the Burgers vector1). According to the optical metallographic examination using an OLYMPUS BX51 microscope, both Ti49.4Ni50.6 and Ti50Ni50 alloys after quenching have a homogeneous coarse-grained microstructure with an average grain size of 200 lm (Fig. 1). In both alloys, impurities in form of
1
According to the calculations on the lattice’s parameter, the Burgers ˚ . While being tested, TiNi can vector of the B2 phase is 2.61 A undergo the B2–B190 phase transformation (during deformation at temperatures lower than Md). During deformation at temperatures above Md, the B2–B190 phase transformation is being suppressed. For the sake of simplicity, the Burgers vector b of the B2 phase of ˚ was used to calculate V [b3] for all conditions. 2.61 A
200 µm
(a)
200 µm
(b)
Figure 1. Microstructure of the quenched alloys Ti49.4Ni50.6 (a) and Ti50Ni50 (b).
(a)
(b)
Figure 2. Microstructure after ECAP processing of: (a) Ti49.4Ni50.6 alloy, (b) Ti50Ni50 alloy.
Ti4Ni2OX spherical inclusions are observed. Since the Ti50Ni50 alloy is a martensitic one (Ms = 78 °C) at room temperature, martensite twins are also observed (Fig. 1b). The Ti49.4Ni50.6 alloy after ECAP processing is in the austenitic state and has a grain size of 0.3 lm (Fig. 2a). The Ti50Ni50 alloy after ECAP is in the martensitic state. From TEM images, the size of austenite grains after ECAP processing cannot be precisely measured due to austenite–martensite phase transformation during cooling after ECAP. Therefore, the size of austenite grains formed during ECAP processing of the Ti50Ni50 alloy was measured as the size of blocks of martensite within individual prior austenite grains. An approximate estimation of the austenite grain size after ECAP processing yields was 0.7 lm (Fig. 2b). A larger size of austenite grains in the ECAP processed Ti50Ni50 alloy as compared to the Ti49.4Ni50.6 alloy can be explained by the fact that the Ti49.4Ni50.6 alloy is an age-hardenable one and the fine particles, which precipitate at the temperature of ECAP processing, can suppress grain growth. Figure 3 demonstrates stress–strain curves of Ti49.4Ni50.6 alloys in CG and UFG states at temperatures of 25°, 150°, 250° and 400 °C. Stress–time plots from strain rate jump
Figure 3. Stress–strain curves from tensile testing of Ti50Ni50 alloy in the CG (a) and UFG (b) conditions.
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
Figure 4. Stress–time curves from strain rate jump testing of Ti49.4Ni50.6 alloy in the CG (a) and UFG (b) states.
testing experiments (where strain rate was changed from 103 s1 to 104 s1) are presented in Figure 4. The results of analysis of stress–strain curves and strain rate jump tests are listed in Table 1, where the average data on yield stress (YS), ultimate tensile strength (UTS), strain sensitivity m, and activation volume V are presented for the Ti50Ni50 and Ti49.4Ni50.6 alloys in the CG and UFG state. During tensile testing at room temperature, a pseudo yield area at rm stresses about 200–400 MPa (depending on the composition and state of the alloy) is observed (Fig. 3). This testifies either to a strain-induced B2–B190 phase transformation (in the Ti49.4Ni50.6 alloy), or to re-orientation of the martensite under applied stress (in the Ti50Ni50 alloy). No pseudo yield area is observed on tensile curves at temperatures above 150 °C. This testifies either to the B2–B190 phase transformation occurring simultaneously with the plastic deformation or to the suppression of the B2–B190 phase transformation. ECAP leads to a considerable increase of the ultimate tensile strength and yield stress of both alloys as compared with the initial quenched state. The UTS and YS of the age-hardenable Ti49.4Ni50.6 alloy in CG and UFG states are somewhat higher compared to those of the Ti50Ni50. Tensile testing of the alloys at room temperature in both CG and UFG states is characterized by a long stage of uniform deformation, which is related to the long-term hardening and twinning of the martensite phase [10,18,19]. The tensile testing at room temperature is finalized by a failure of samples without localization of plastic deformation and without formation of any neck. Strength and ductility of both alloys tend to decrease with increasing testing
101
temperature. Decrease of ductility with increasing temperature can be related to the suppression of the B2–B190 phase transformation, reduction of the uniform deformation and its faster localization. The strain rate sensitivity m increases with increasing testing temperature in both alloys in CG and UFG states. Depending on the composition and testing temperature, the m parameter in the UFG state is by a factor of 1.5–3 higher than the m parameter of the alloys in the CG state. The m parameter of the Ti50Ni50 alloy is also somewhat higher than m of the Ti49.4Ni50.6 alloy at all testing temperatures. The parameter m measured at RT has a considerably small value (0.002–0.007). This may be associated with the fact that twinning plays a significant role in the deformation of TiNi. It should be also noted that twinning has a negligible strain rate sensitivity, and a material that deforms primarily through twinning would be expected to have an abnormally low m compared to a material that deforms from dislocation mechanisms at any given temperature [21]. The activation volume V in the CG and UFG alloy Ti50Ni50 and the UFG alloy Ti49.4Ni50.6 grows with increasing temperatures to the maximum value (at T = 150°– 250 °C) and then drops at higher temperatures. In most of the cases (excluding the CG Ti49.4Ni50.6 alloy), a temperature increase up to 400 °C results in reduction of activation volume. In the CG Ti49.4Ni50.6 alloy, activation volume grows with increasing temperatures up to 400 °C. The parameter V is determined by a relationship between the values T, m and r (Eq. (2)). The strain rate sensitivity m increases with increasing testing temperature for all the alloys in all states. However, for the Ti50Ni50 alloy the parameter m grows with increasing temperature faster than for the Ti49.4Ni50.6 alloy. The dependence of the flow stress r on the testing temperature T is ambiguous and differs for alloys/state. All this accounts for the complexity of the dependence of the V-value on the testing temperature T for different alloys/states. The growth of the activation volume simultaneously with increasing testing temperature for T = 150°–250 °C could be rationalized based on the fact that tensile testing of both alloys at room temperature occurs in the martensitic state, whereas austenite–martensite phase transformation is suppressed at elevated temperatures.
Table 1. Results of mechanical testing of Ti50Ni50 and Ti49.4Ni50.6 alloys in CG and UFG states. d (%)
m
V, b3
823 494 471 318
59 36 35 17
0.004 0.004 0.01 0.03
169 419 461 119
870 584 598 703
1012 831 720 736
38 27 19 4
0.007 0.01 0.02 0.05
52 154 77 30
25 150 250 400
477 552 597 543
945 768 792 626
30 15 15 11
0.002 0.003 0.004 0.004
269 319 284 604
25 150 250 400
1160 906 948 1020
1256 1159 1094 1091
18 8 10 8
0.006 0.007 0.007 0.012
121 119 105 99
Alloy/state
Temperature (°C)
YS (MPa)
Ti50Ni50/CG
25 150 250 400
430 377 364 287
Ti50Ni50/ECAP
25 150 250 400
Ti49.4Ni50.6/CG
Ti49.4Ni50.6/ECAP
UTS (MPa)
102
D.V. Gunderov et al. / Scripta Materialia 102 (2015) 99–102
Reduction of activation volume in most of the cases (excluding the CG Ti49.4Ni50.6 alloy) with a temperature increase up to 400 °C can be related to the enhanced contribution of diffusion and intergranular sliding at this temperature [20]. The different forms of the V(T) dependence for the CG and UFG Ti49.4Ni50.6 alloy with increasing temperature up to 400 °C could be explained by the fact that the kinetics of decomposition of solid solution and formation of precipitates in the UFG Ti-Ni age-hardenable alloys is significantly different from that in their CG counterparts [15], and these processes have a noticeable effect on deformation. However, to account for the observed V(T) dependence for CG and UFG TiNi with different compositions, additional studies are definitely required. It is known that activation volume within a range of 1– 10 b3 points at diffusion and grain boundary sliding as the controlling deformation mechanisms [13,17]. Annihilation of dislocations at grain boundaries dominates plastic deformation if the V value is within a range of 10–100 b3 [17]. In case the V value is within a range of 100–1000 b3, the overcoming of the forest junctions by gliding dislocations is considered as the controlling deformation mechanism [13,17]. Therefore, according to the V value, within the studied range of temperatures in the CG state of TiNi (V 100–400 b3), thermally activated overcoming of forest junctions by moving dislocations controls thermally activated slip. In case of the UFG TiNi alloys (V 20– 120 b3), plastic deformation is mainly controlled by annihilation of dislocations along grain boundaries. In summary: ECAP leads to a considerable increase of the UTS and YS of both TiNi alloys as compared with the initial state. The UTS and YS of the age-hardenable Ti49.4Ni50.6 alloy in CG and UFG states were somewhat higher compared to those of the Ti50Ni50. Tensile testing of the alloys at room temperature is characterized by a long stage of uniform deformation, a failure of samples took place without localization of plastic deformation. With increasing testing temperature ductility decreased, that related to the suppression of the B2–B190 phase transformation and faster deformation localization. The UFG TiNi alloys have higher strain rate sensitivity and lower activation volume compared to their coarsegrained counterparts in the temperature range of 20– 400 °C. Strain rate sensitivity tends to increase with increasing temperature in CG and UFG TiNi alloys, whereas the effect of temperature on activation volume is ambiguous. This is accounted for by the complexity of phase transformations in the material at increasing testing temperature T (martensite–austenite transformation, solid solution decomposition, aging), which has a complex effect on flow stress.
D.V. Gunderov acknowledges gratefully funding through the Russian Government Program of Competitive Growth of Kazan Federal University. G.I. Raab acknowledges gratefully financial support by the Russian Ministry of Education and Science of Russia through project No. 1.729.2014R.
[1] V. Brailovski, S. Prokoshkin, P. Terriault, F. Trochu (Eds.), Shape Memory Alloys: Fundamentals, Modeling and Applications, Ecole de Technologie Superieure (ETS), Universite du Quebec, Montreal, Canada, 2003, 851p. [2] K. Otsuka, X. Ren, Prog. Mater. Sci. 50 (2005) 511. [3] L. Mishnaevsky Jr., E. Levashov, R.Z. Valiev, J. Segurado, I. Sabirov, N. Enikeev, S. Prokoshkin, A.V. Solov’yov, A. Korotitskiy, E. Gutmanas, I. Gotman, E. Rabkin, S. Psakh’e, L. Dluhos, M. Seefeldt, A. Smolin, Mater. Sci. Eng. R 81 (2014) 1. [4] T. Waitz, V. Kazykhanov, H.P. Karnthaler, Acta Mater. 52 (2004) 137. [5] B. Kockar, I. Karaman, J.I. Kim, Y.I. Chumlyakov, J. Sharp, C.J. Yu, Acta Mater. 56 (2008) 3630. [6] M.L. Young, M.F.X. Wagner, J. Frenzel, W.W. Schmahl, G. Eggeler, Acta Mater. 58 (2010) 2344. [7] V.G. Pushin, R.Z. Valiev, Y.T. Zhu, D.V. Gunderov, T.E. Kuntsevich, N.I. Kourov, A.N. Uksusnicov, L.I. Yurchenko, Mater. Trans. 47 (2006) 694. [8] A.V. Sergueeva, C. Song, R.Z. Valiev, A.K. Mukherjee, Mater. Sci. Eng. A 339 (2003) 159. [9] V.G. Pushin, V.V. Stolyarov, R.Z. Valiev, T.C. Lowe, Y.T. Zhu, Mater. Sci. Eng. A 410–411 (2005) 386. [10] R.Z. Valiev, D.V. Gunderov, A.V. Lukyanov, V.G. Pushin, J. Mater. Sci. 47 (2012) 7848. [11] I.Yu. Khmelevskaya, S.D. Prokoshkin, I.B. Trubitsyna, M.N. Belousov, S.V. Dobatkin, E.V. Tatyanin, A.V. Korotitskiy, V. Brailovski, V.V. Stolyarov, E.A. Prokofiev, Mater. Sci. Eng. A 481–482 (2008) 119. [12] I. Karaman, A.V. Kulkarni, Z.P. Luo, Philos. Mag. 85 (2005) 1729. [13] P. Rodriguez, Metall. Mater. Trans. A. 35 (2004) 2697. [14] Y. Zhao, Y.T. Zhu, E.J. Lavernia, Adv. Eng. Mater. 12 (2010) 769. [15] E.A. Prokofiev, J.A. Burow, E.J. Payton, R. Zarnetta, J. Frenzel, D.V. Gunderov, R.Z. Valiev, G. Eggeler, Adv. Mater. Eng. 12 (2010) 747. [16] D. Gunderov, A. Lukyanov, E. Prokofiev, A. Churakova, V. Pushin, S. Prokoshkin, V. Stolyarov, R. Valiev, Mater. Sci. Forum 738–739 (2013) 486. [17] D. Caillard, J.L. MartinThermally Activated Mechanisms in Crystal Plasticity, Pergamon Materials Series, vol. 8, Elsevier, Oxford, United Kingdom, 2003. [18] J.X. Zhang, M. Sato, A. Ishida, Acta Mater. 54 (2006) 1185. [19] J.X. Zhang, M. Sato, A. Ishida, Acta Mater. 49 (2001) 3001. [20] M. Kawasaki, T.G. Langdon, J. Mater. Sci. 42 (2007) 1782. [21] C.H.J. Davies, Mater. Sci. Forum 539–543 (2007) 1723.