Materials Characterization 106 (2015) 218–225

Contents lists available at ScienceDirect

Materials Characterization
journal homepage: www.elsevier.com/locate/matchar

Effects of strain rates on deformation twinning behavior in α-titanium
Tongbo Wang, Bolong Li ⁎, Mian Li, Yingchao Li, Zhenqiang Wang, Zuoren Nie
College of Material Science and Engineering, Beijing University of Technology, Beijing 100124, China

a r t i c l e

i n f o

Article history:
Received 11 December 2014
Received in revised form 22 April 2015
Accepted 12 May 2015
Available online 22 May 2015
Keywords:
Commercial pure titanium
High strain rate
Deformation twins
EBSD

a b s t r a c t
The deformation twins of commercial pure titanium were investigated under the split Hopkinson pressure bar
compression with different strain rates. It was demonstrated that the types of twins induced by high-speed comn
o n
o
n
o n
o
pression included 1122 , 1124 contraction twins and 1012 , 1121 extension twins. Furthermore, the
quantities of the deformation twins at the high strain rates were more pronounced than those at the low to men
o
n
o
dium speed deformation. The content of 1122 twin increased with the strain rate, while 1121 twin's quantity decreased slightly with the increasing strain rate. The semi-quantitative relationship between twin density
and strain rates was derived and discussed, which indicated that twin density was nonlinearly proportional to
the strain rate.
© 2015 Elsevier Inc. All rights reserved.

1. Introduction
The deformation behavior of hexagonal close-packed (hcp) metallic
materials is complicated due to the activation of both different slip systems and twinning systems [1]. In the past several decades, the occurrence of deformation twinning in hcp metallic materials under low
speed deformation has been widely reported [2–5]. In terms of the application of hcp metallic materials in high strain rate circumstances
[6], much more attention should be paid to its high-speed deformation
behavior. As is known [7,8], α-titanium is a typical hcp metallic material, whose ratio of c to a is 1.587. Owing to its low symmetry, αtitanium's deformation is governed by both dislocation slipping and
twinning. In fact, the plastic deformation of α-titanium is caused mainly
by dislocation slipping. Meanwhile, twinning can transform the grain
orientation with a lower Schmid factor to the benefit orientation for
dislocation slipping. Therefore, high-speed deformation twinning behavior becomes a significant part of α-titanium's plastic deformation
mechanism [9].
For hcp structural α-titanium, the type of dislocation slipping includes baN and ba + cN type. The main dislocation slip systems consist
n
oD
E
D
E
n
oD
E
of 1100 1120 , f0002g 1120 and 1011 1123 , whose critical
shear stresses are 30 MPa, 150 MPa and 125 MPa, respectively [10].
On the basis of less independent slip systems and lower stacking fault
energy, twins become the main factor for further deformation by
adjusting the hard orientation to the soft one. Deformation twins are

⁎ Corresponding author at: College of Material Science and Engineering, Beijing
University of Technology, China.
E-mail address: blli@bjut.edu.cn (B. Li).

http://dx.doi.org/10.1016/j.matchar.2015.05.016
1044-5803/© 2015 Elsevier Inc. All rights reserved.

formed by the mutual shearing of parallel lattice plane, and they belong
to an important mechanism for plastic deformation. In the low-speed
n
o n
o
plastic deformation, the main twins include 1012 , 1121 and
n
o
n
o
1122 . Among these, 1012 extension twin is one of the most active
n
o
twins induced by deformation at ambient temperature, and 1122
twin belonged to the contraction twins [10,11]. Rosi [12] has suggested
n
o
n
o n
o
that the deformation of 1100 slip system and 1012 , 1121 ,
n
o
1122 twinning system should be more active in the quasi-static plastic deformation.
Jonas et al. [13] investigated the twinning behavior of commercial
pure titanium at the quasi-static compression, suggesting that the critin
o
cal strain for twinning was sensitive to the strain rate. The 1122 twin
n
o n
o
is the most popular type, and a large number of 1122 – 1012 twins
n
o n
o
and a few 1012 – 1122 twins were generated during the quasistatic compression. Moreover, the presence of low Schmid factor twins
was established as well as the absence of potential high Schmid factor
twins, and the high Schmid factor twins needed the dislocation slipping
for further deformation. A. Serra and his coworkers [14] reported that
n
o
n
o
1121 and 1122 twins had blocked or synergistic effects on the different dislocation slip systems, and pointed out that the dislocationtwin structure induced by deformation greatly contributed to the
strength of the material, which was consistent with other results [11,
15,16]. Research on dislocation-twin structure became a significant
issue, and further study [17] found that the interaction between
n
o
1121 and (a + c) type dislocation slipping was beneficial to the strain

2. 2. IPF map. Similar to the IPF map. It should be emphasized that there are few twins inside the grains. High-speed compression The device employed to realize the high-speed deformation was the split Hopkinson pressure bar (SHPB). and the roughness of specimens' surface was 1.1. the strain rate. the contacting surface between the sample 2.11 was performed on JEM-2010F transmission electron microscopy (TEM).15 Bal. copper [19. the content of αtitanium's twin can increase under high-speed deformation. α-Fe [18]. The electro-polishing was performed at 35–40 V. 0. HKL Channel 5 software was used to process the data. During high-speed compression. and the Schmid factor of the different grains.165 atm—3500/s. Initial microstructure and dynamic strain rate The IPF map can illustrate the microstructure and orientation distribution of the grain.6 mm. IPF map and misorientation distribution of as-annealed commercial pure titanium. Its microstructure consisted of equiaxed grains with an average size of 50 μm. The cylindrical specimens with the dimensions of Φ4 × 4 mm were cut from the as-annealed commercial pure titanium through electrical discharging cutting. where the data with misorientation of less than 2° are excluded. . Besides. Samples were mechanically polished to a thin sheet. and the backscatter signal data were obtained by Nordlys collection probe. and aluminum [21] could generate twins during high-speed deformation.1. The cylindrical specimen and the pressure bars must be coaxial. In addition. These samples were examined in a Quanta FEG 650 scanning electron microscope.% 0. indicating a recrystallized microstructure. and no twins could be seen. Fig. Obviously. In the present paper. Results and discussion 3. and selected area diffraction patterns were analyzed at the twinning boundaries. A disk with a diameter of 3 mm was twin-jet electro-polished of in a 60% HClO4 50 ml-CH3(CH2)3OH 350 ml-CH3OH 600 ml solution at 248 K and 20 V. 1 shows the IPF map and the misorientation distribution of as-annealed commercial pure titanium.05 0.T. whose chemical composition is shown in Table 1. which was a steel tube with a thickness of 3. the effects of the strain rate (ranging from 1500/s to 3500/s) on deformation twinning behavior of α-titanium were investigated.04 0.23]. and the stress could be calculated by the signals.6. the effects of strain rates (≥1000/s) on the type and content of the twin need more detailed study and discussion. Element Fe C N H O Ti wt. the strain was limited to 10% by the blocking ring. The striker bar has a length of 200 mm. in order to analyze the misorientation. The strain. Moreover. the misorientation distribution clearly indicated that no preferred orientation was induced by twinning in as-annealed α-titanium. and the relationships between them were 0. Microstructural analysis hardening. 3. 2. The different strain rates were attained by adjusting the gas pressure to the striker bar.20].2. 219 and the bar was coated with Vaseline to minimize the transverse friction. 1. the type and content of the twin. the annealed microstructure reveals the equiaxed grains with an average size of 50 μm. Fig. but it was difficult to form twins during the low-speed deformation. the error of dimension was controlled in the range of 0. As shown in Fig. which can enhance particular texture [22. and 0.1. 120 s.2. 2.015 0. 2.30 0. TEM Microstructure observation of commercial pure titanium impacted at the strain of 0.11 atm—2500/s. the distribution profile of the large-angle and small-angle grain boundaries in annealed titanium was consistent with the theoretical curve and without the distinct peaks. the paper makes an effort to provide an initial theory for the high-speed deformation mechanism of hcp materials. EBSD As-impacted specimens were coarsely grinded and electro-polished for Electron Backscatter Diffraction (EBSD) testing. 1(a). Experimental procedure The as-annealed commercial pure titanium (grade 1) was selected as a model material. 1(b) presents the misorientation distribution. and the electropolishing liquid was a solution of 60% perchloric acid (6 ml) and glacial acetic acid (94 ml). Wang et al. Sub-structures were obtained at 200 keV by the standard TEM techniques. The input and output bars' signals could be obtained by the strain gauges.5 mm. / Materials Characterization 106 (2015) 218–225 Table 1 Chemical composition of commercial pure titanium (wt%).07 atm—1500/s. In particular. It aimed to elaborate on the relationship between the twinning behavior and the strain rate. However.02 mm.2. which was made up by two highstrength steel bars with the diameter of 14. and ambient temperature. Fig.

so the amount of 1121 twin was relan o tively few. Therefore. 65°. This may be attributed to the strain energy of various twins. the twinning behavior appeared inside the grains. The IPF maps of α-titanium impacted at the strain rate of 1500/s. f). and 80–90° axes correspond to 1210 orientation. inD E D E D E D E cluding 35° 1100 . 3 shows the IPF and the twinning map of as-impacted α-titanium. Obviously. so twinning was more active during highspeed deformation. Similarly. the content of the twin increased at the high-speed deformation. and 77° 1100 (the error was ±5°).67%. Importantly. and all index rates are up to 90%.66. In other words.065–0. Wang et al. which could be attributed to the formation of the incidental dislocation boundaries and the geometrically necessary boundaries by the dislocations' interaction. The inter-granular misorientation distribution EBSD testing was conducted with an area of 400 × 400 μm. [24]. the figure shows the curves with different slopes. and 1124 twins. In addition.5. the type and content of the twin were determined by inputting the axis-angle couple. there were three distinct peaks on the 35°. 2500/s and 3500/s. e). the nucleation of the twin started from the grain boundary. Deformation twinning response to the strain rate 3. Due to the interaction among the grains. The content of the twin was counted by the Image-pro Plus software. the quantity n o n o of 1122 compression twin was the largest. Linear fitting to the strain-time curve was performed. the lower the energy of the twin was.638. 64° 1100 .2. d. The total twin's density was 27. As shown in Fig. Fig.175. 1121 twin's n o shear strain was up to 0. However. The preferred misorientations were generated during the high-speed compression. 28. which was proportional to the square of twin strain. and 87. However. 2500/s and 3500/s corresponded to Fig. Owing to the shorter deformation time during the high-speed deformation. the . As shown in Table 3. n o n o 1121 . Fig. 3. it D E D E was indicated that the main twins included 35° 1100 . 3 n o n o (b. n o n o 1012 . / Materials Characterization 106 (2015) 218–225 Fig. 2502. The slopes were 1569. These data were rounded to 1500. and 1124 formed at the high-speed deformation. 5 shows the densities of different twins.62%. the quann o n o n o tity of 1122 and 1012 twin was larger. 3 (a. The axis-angle regimes D E n o D E n o D E include 35° 1100 — 1121 .17.225 and 0. 3. the rotation axis distribution of the misoriD E entation peaks 30–40° and 60–70° had the 1100 preferred orientaD E tion. 64° 1100 — 1122 .2. the twin density was 17–18.2. respectively.220 T. As reported in Ref. 64° 1100 . The recrystallized grains were fragmented by impacting from 4. Combined with the deformation structure and the misorientation distribution. 1122 . c. As can be seen.2. induced by the high-speed deformation at the strain rate of 1500/s. most deformation twins distributed in the grain interior and crossed through the whole grain. the content of the low angle boundaries (2°–15°) underwent an obvious increase after the high-speed compression. the 77° 1100 twin could form during the medium-speed deformation.60 mm at the strain rate of 1500/s. The rotation axis distribution of the misorientation peak was reflected by the inverse pole figure for Z of twins in titanium. the shear strain of 1122 n o and 1012 was 0. and 3500/s. it was easier to cause stress concentration for the nucleation of twin in the grain boundary. The type and content of the deformation twins at different strain rates According to the coincidence lattice relationship. 2. 85° 1210 . step size of 1. respectively [10]. According to the misorientation distribution and the preferred axis orientation. Clearly. D E D E and 85° 1210 . Fig. and the slopes stand for the strain rates. In conclusion. so the average strain rates were decided as 1500/s. Additionally. It could be convinced that the twins marked 1121 . Curve for strain and time to determine strain rate. Namely. followed by 1012 .5° positions at the misorientation distribution profile for all impacted samples.45 and 3456. The twin density of 1122 increased when the strain rate increased from 1500/s to 3500/s at the strain of 10%. 2 shows the curve for the strain and time.30%. 85° 1210 — n o D E n o 1012 . 2500/s and 3500/s. and the strain can be calculated by signals obtained from the strain gauges. the formation of the twin needed less micro-spacing to start with than dislocation slipping. and 77° 1100 — 1124 . 4. The types and distributions of the twins are shown in Fig. and the results are presented in Table 2. Suwas and his coworkers [22].00 mm to 3.5% at the strain rate of 10− 5/s and the strain of 0. 2500/s. 2500 and 3500.18. the appearance of peak may be due to the twinning behavior. Additionally. reported by S. the dislocations' proliferation and long range migrations became more difficult. respectively. and 31.1. the easier the forman o tion of the twin became.

it could be strongly indicated that the twin density of ergy to form during the deformation. the density of n o 1121 twins induced by the deformation at the strain rate of 5400/s n o 1121 decreased slightly. the 1122 twin had lower strain en- was 2. 3. / Materials Characterization 106 (2015) 218–225 221 Fig. Additionally. . n o twin density of 1121 declined slightly. This phenomenon may be due to the strain n o energy to form twinning. Wang et al.5%. IPF map and twinning map of commercial pure titanium impacted at strain rate of 1500/s. namely. 2500/s and 3500/s. Thus.T.

As shown above. we can derive the partial differential of the dislocation density against the strain both in the micro and macro aspects. the twins. considering the factor of the grain size. 3). and activation energy for diffusion respectively. Therefore. Bouaziz and Guelton [26] raised up the partial differential (Eq. the value-added and annihilation of the dislocations:       pffiffiffiffiffiffi ∂ρ M 1 1 F −mr t þ þ k ρ0 −f ρ0 : ¼ 1−F b d 2eð1− F Þ ∂ε mic simulate formulas (2. The Johnson–Cook mode reflected the constitutive relationship of the titanium's high speed deformation. R. T. (3) in which A. However. gas constant. (4). (1).3. Therefore.222 T. and the others were constants. and C were constants:  pffiffiffi σ ¼ σ 0 þ αMGb  ρ  n σ ¼ ðA þ Bε Þ 1 þ C ln ε The partial differential of the dislocation density on the strain was solved in the macro-aspect.30% 31. and Q represent the strain rate.60 at strain rates of 1500/s. the yield stress (σ0). B. 4. Wang et al. For this. Semi-quantitative analysis of relationship between the twin density and the strain rate As shown in Eq. as shown in Eq. . and it increased with the strain rate. 3. and organize as the Eq.   Z ¼ε exp  Q RT ð1Þ  where ε. the Zener–Hollomon equation [24. the Zener– Hollomon equation could explain why the total contents of the twins increased. we can suppose ð2. the Z parameter increased with the strain rate at room temperature. the grain size and the twins' thickness respectively. (5). d and e stood for the twin density. presented in Eq.       2nB 1 þ C ln ε  ∂ρ ¼ ðA þ Bεn Þ 1 þ C ln ε −σ 0 εðn−1Þ : 2 2 2 2 ∂ε mac α M G b   The hcp structural material's deformation mechanism was dislocation slipping and twinning. As was known. it did not give the quantitative or semi-quantitative relationship between the twin density and strain rate. which was to suppress the dislocation activity and stimulate the twinning. Contents of various twins.62% 28.25] gives a reference value to display the relation between the strain rate and the experiment temperature by citing the Z parameter. F. temperature. (6)) of the dislocation density on the strain in the micro-aspect. which incorporated the flow stress (σ). 5. 3Þ  ð5Þ Fig. (2). Intergranular misorientation distribution of commercial pure titanium impacted from 4.2. O.67% ð6Þ In the above equation. / Materials Characterization 106 (2015) 218–225 Fig. 2500/s and 3500/s. The mechanical behavior was assumed to obey the classical relation between the flow stress and the total dislocation density in Eq. It was reported that the Z parameter of the commercial pure titanium was 170 kJ/mol. the microstructure and macroscopic properties of metallic materials corresponded to each other. Obviously.00 to 3. and the dislocation density. so the formula was attempted to be derived in the following section. State As-annealed 1500/s 2500/s 3500/s Stwins/S ≈0 27. ρ¼ 1    2 ðA þ Bεn Þ 1 þ C ln ε −σ 0 :  2 α 2 M2 G2 b ð4Þ Table 2 The total twin density of α-titanium impacted at different strain rates (Stwins/S).

the quantity of the small angle boundary increased rapidly with the true strain increasing from 0.11. Namely. ð9Þ that Eq. ε ¼ 1−   G ε. the semiquantitative relationship meets the function. Eq.22 at the strain rate of 1500/s. ε   0. ν. The intergranular misorientation distribution at the strain of 0. In addition. t Þ ¼ λðA þ Bε n Þt 2 −λσ 0 t þ ω based on   σ ¼ ðA þ Bε n Þ 1 þ C ln ε  so. and ω are constants. ∂G ¼ λ½2ðA þ Bε n Þt−σ 0  ¼ λ½2σ−σ 0 ≥0: ∂t Fig.175 o n o 1121 0. the more twins formed. Gðε.225 then. and need to be further fitted. and the peaks at the misorientation of 35°. Eq. 6. ε ¼ λðA þ Bεn Þ 1 þ C ln ε −λσ 0 1 þ C ln ε þ ω    ordering. As shown above. / Materials Characterization 106 (2015) 218–225 Supposing. (8) is also a monotonically increasing function from zero to positive infinity because C is a positive constant. and the calculation process is as follows. Additionally. By being sorted out and simplified. (b) SADP showing 1122 twin relation. (7) is an increasing function for the strain rate. Eq. However. The contribution of twins for the high-speed deformation   λðA þ  Bεn Þ νεð1−nÞ : 2   1 þ C ln ε −λσ 0 1 þ C ln ε þ ω   ð7Þ In the above equation. the twin density function of the strain and strain rate was shown as Eq. Sub-structures of commercial pure titanium impacted at strain of 0. 65° and 87. As shown in Fig. n Twin system Twin shear 1012 0. Wang et al.638 223 n 1122 o 0. (5) was equal to (6).3. (9) is a monotonically increasing function.T. the monotonic nature should be discussed. (a) 1122 twin.   F ε.5° still occurred as well. Table 3 Twin shear associated with different twin systems. which is determined by the fitted results of commercial pure titanium.11 and 0. 7. The formula was simplified because the coefficients of ε and ε were determined by the material constants above. Meanwhile. (c) dislocation slipping and (d) dislocation boundaries. (7). 6.22 (1500/s). ε ¼ 1− 3. λ.11 to 0. Based on above.    2   G ε. the larger the strain rates were.  t ¼ 1 þ C ln ε ð8Þ then. . the strength of the peaks at n o n o Fig.225 n 1124 o   νεð1−nÞ F ε.

which was related to the further deformation assisted by the twins. while the peak at 87. d). 8.11 to 0. In addition. The incidental dislocation boundaries and the geometrically necessary dislocation boundaries formed by the dislocation slipping. The dislocation lines got together towards the boundary to form the geometrically necessary boundaries (GNBs) and the incidental dislocation boundaries (IDBs). Schmid factor map of commercial pure titanium impacted at different strain rates of (a) starting material. Therefore. 7(b) shows the diffraction pattern. 7(a) were the twins with the width of 0.22. Fig. The twins play an important role in adjusting the crystal orientation and releasing the stress concentration to stimulate further dislocation Fig. as shown in Fig. . These dislocation boundaries made up the smallangle grain boundaries demarcated by EBSD.22. and the twinning width in the TEM microstructure was consistent with the one in the IPF map. (b) 0. The characteristic microstructures included the twins. n o which reveals the 1122 twin relation. so the quantity of the small angle boundaries increased with the increasing of the strain. Wang et al. Straight GNBs with the consistent orientation were the low-energy dislocation structures composed by the high density dislocation wall. There were dislocation line structures in the expected dislocation slipping area. The formation of the incidental dislocation boundary was formed by the interaction between the simulated slipping dislocations and the forest dislocations. 6 indicates that there was the first contraction twin followed with the extension twin with increase in the strain. / Materials Characterization 106 (2015) 218–225 35° and 65° decreased sharply. Fig.5° increased significantly with the strain increasing from 0. The geometrically necessary boundaries formed to coordinate the deformation of the adjacent cell blocks. The main sub-structures shown in Fig. The microstructure of the commercial pure titanium impacted at the strain of 0. the Schmid factor distribution of the main dislocation slip systems should be discussed in the following paragraph.224 T. as shown in Fig.11 was analyzed by transmission electron microscope (TEM). 7 (c. 7.485 μm. the dislocation tangle and the dislocation boundaries.11 and (C) 0. the dislocation slip is stimulated during the period of deformation with the coordination of the twin to form the dislocation cell. The selected area electron diffraction (SAED) was applied to determine the twinning relation.

Daob. B. Mater.B. 71 (2014) 293–305. Yoo.L. the increase of n o 1012 twin makes it easier to be stimulated. Salem. 55 (2006) 319–322. J. Mater. G. Suresh. the grains around the 1012 twin had n oD E the high Schmid factor for 1122 1123 dislocation slipping. Mater. [9] S. Microstructure and property of nitrogen-alloyed high manganese austenitic steel under high strain rate tension. Phys. P. A39 (2008) 810–843. S. 5 (1957) 337–339. Eng. [19] Y. the high-speed deforman o n o tion twins include 1122 . Trans. A 33A (2002) 809–812.A. Hemker. F. Trujilloa. Initiation and accommodation of primary twins in high-purity titanium. Sci. Eng. Trans. R.M. C. Wang et al. Acta Mater. Gray. Semiatin. M. Sci. Salem.N. Metall.L. R. 2004 (1889–1892) 155–156. Phys. Hollomon. The grains with the high Schmid factor should be paid much attention n o to. X. [14] A. Degrieck.J. Bouaziz. Zener. [13] Hong Qin. Mater. [22] N.D. [17] Shinji Ando. The formation of the twin contributed mostly to the ba + cN slipping for as-impacted α-titanium. Metall. A 541 (2012) 190–195. Mater. Mater. L.Y. the type and content of the high-speed deformation twins. [5] A. Microstructure of adiabatic shear bands in Ti6Al4V. Study of 1121 twinning in a-Ti by EBSD and Laue microdiffraction. Deformation behavior of commercially pure titanium at extreme strain rates. Molecular dynamics simulation of (c + a) dislocation core structure in hexagonal-close-packed metals. Jonas. S. Eng. n oD E (3) For the 1011 1123 dislocation slip system.A. Slip. Sci. 84 (2013) 41–47. W. Twins as barriers to basal slip in hexagonal-closepacked metals. Rosi. Y. However. Strain hardening due to deformation twinning in α-titanium: constitutive relations and crystal-plasticity modeling. 53 (2005) 3495–3502. Rajeev Kapoor. C. K. Sci. Ramesh. Rabet. Kevin J. 47 (1999) 3705–3720. Philip Eisenlohr. J.R. N.R. [23] D.C. Bacon. Peirs. n o [24] Feng Xu. Kalidindi. Acta Mater. 1124 contraction twins and n on o 1012 . D. [11] Leyun D Wang. Mater. S.H.F. Cerretaa. S. 1124 deformation twinning in pure Ti during dynamic plastic deformation. [21] L. Trans. and fracture in hexagonal close-packed metals.P. K. [8] X. Yilong Bai. Els-Botes. there was less independence slip system to support further deformation. J. Charact. The dislocation slip planes included n o n o 1100 . 61 (2013) 844–879.T. Scr. Williams. I. Frank. Trans. Acta Mater. Microbands and shear-related microstructural phenomena associated with impact craters in 6061-T6 aluminum. Wenjin Liu. Marc Andre Meyers. Acta Mater. A 44a (2013) 3664–3674. [25] C. et al. Twin intersections in titanium. 75 (2013) 79–92. Metall. 1122 twin occurs most frequently and increases gradually. L. L. 51371013. Pond. Korzekw. Jinghua Zhang. Xiyan Zhang. Peng. A 501 (2009) 207–219. EBSD characterization of twinning in cold-rolled CP-Ti. The high-strain-rate response of alpha-titanium: experiments. Chichili. Coghec. Steven Celotto. Mater. 225 References [1] Dipankar Banerjeea.H. Cheng. A 216 (1996) 69–79. M. A 12A (1981) 409–418. . Semiatin. 96 (2014) 206–212. L. [7] Robert C. but the total content of the twin increased sharply after high-speed compression. Interfacial deformation mechanisms in hexagonalclose-packed metals. Thus. Hideki Tonda. Sci. Thomas Bieler. In order to investigate the contribution of the twin to the dislocation slipping. Dynamic mechanical behavior of pure titanium. Dai. Pond. Guo. Charact. G. Verleysen. the quantity of 1012 twin increased with the strain n o from 0. 59 (2011) 3431–3446. A 33A (2002) 801–807. Appl. The Schmid factor distribution of the main dislocation slip systems is shown in Fig. n o (2) At the strain rate ranging from 1500/s to 3500/s. Ramesh. Guelton.G. 1101 and {0002} planes.F. Z. John J. / Materials Characterization 106 (2015) 218–225 slipping. Woudberg. [20] E. 4. McGrath. Trans. S.L. This may be attributed to the twinning energy and short the time of deformation. A 492 (2008) 255–260. E Rozaliya Barabash. Technol. Charact. 8. [16] A.E. [15] A.C. Xua. [12] F. P. Shear localization in dynamic deformation: microstructural evolution. On the basis of the results. Bassim. Takushi Gotoh. Trans. [26] O. [18] Fu-cheng Zhang. Lu. Eng. Serra. but the content of n o 1121 twin decreases slightly.P. twinning. the Schmid factor was calculated by the HKL Channel 5 software. Metall. 46 (1998) 1025–1043. deformation mechanisms and modeling. 1121 extension twins in α-titanium. Mech. Conclusions In the present work. and the ba + cN type could be stimulated as well. Tirry.D.T. the higher Schmid facn oD E tor was beneficial to stimulate the a-type 1100 1120 and (a + c)n oD E type 1122 1123 slip system. Mater. which indicated that the dislocation slipping system is difficult to be stimulated. Satyam Suwas. Hemker. Niou. J.A. Strain hardening due to deformation twinning in α-titanium: mechanisms. Effect of strain-rate upon the plastic flow of steel. Metall. Mechanical characterisation of the effect of various forming processes applied to commercially pure titanium. Acta Mater. Solids 52 (2004) 1889–1909. Li. Haitao Ni. Mater.11 to 0.J. Modelling of TWIP effect on work-hardening. Xiyan Zhang. [3] Qiang Li. Li. Trans. W. S. Adiabatic shear localization in αtitanium: experiments. Fidder. K.J. Mater. Mater. Metall. Process. [10] M.Y. Shen. Mechanical properties and deformation mechanisms of a commercial pure titanium. In addition. Qing Liu. Liang Qiu. the twin density function of the strain and the strain rate is established to describe the semi-quantitative relationship between the twin density and the strain rate. Mater. Nemat Nasser. conclusions can be drawn as follows: (1) Similar to the low-speed deformation. Acknowledgements This research was supported by the National Natural Science Foundation of China (NSFC). Strain rate sensitivity of Cu with nanoscale twins. A 33A (2002) 823–829. [2] Deepak R. and the Schmid factors have been characterized by the EBSD analysis method.T. Metall. Amin-Ahmadi. C. Mater. H. Doherty. Mater. n o the evolution of the twin suggested that the presence of 1012 twin n o n o was established as well as the absence of 1121 and 1122 twins during the further deformation.22. Yang-zeng Zheng. D. Perspectives on titanium science and technology. Mater. 15 (1944) 22–32. 6. Acta Mater. Alicia Ayala. [6] J. Obviously. J. Zhang. Rui-dong Fua. [4] Yongbo Xu. modeling and microstructural evolution. Project No. Duan. A 37A (2006) 259–268.R. Murr. Mater. A 319–321 (2001) 246–249. In the first row of Fig..-S. Gurao. Y. In addition.T. Xu. so the further deformation needed to be stimulated by twinning. Dougherty. Acta Mater. Kalidindi.K. The darker color stands for the grain with smaller Schmid factor. Chichili. J.J. Schryvers.G. The influence of microstructure on the mechanical response of copper in shear. D. Eng.