- Email: [email protected]
Hot deformation behavior and dynamic recrystallization of melt hydrogenated Ti-6Al-4V alloy
Journal of Alloys and Compounds 728 (2017) 709e718
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Hot deformation behavior and dynamic recrystallization of melt hydrogenated Ti-6Al-4V alloy Xuan Wang, Liang Wang*, LiangShun Luo, XiaoDong Liu, YingChun Tang, XinZhong Li, RuiRun Chen, YanQing Su**, JingJie Guo, HengZhi Fu National Key Laboratory for Precision Hot Processing of Metals, School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 April 2017 Received in revised form 3 August 2017 Accepted 5 September 2017 Available online 8 September 2017
The effect of hydrogen on hot deformation behavior of Ti-6Al-4V alloy was investigated. Ti-6Al-4V alloy was hydrogenated by melting alloy in gas mixture of hydrogen and argon (melt hydrogenation). Experimental results of hot compression at same strain rate showed hydrogen decreased flow stress at higher deforming temperature, which was attributed to hydrogen induced dislocation movement and dynamic recrystallization (DRX). At lower temperature, peak stress firstly decreased and then increased with increasing hydrogen content. Hydrogen decreased the peak stress and improved the hot workability of alloy deformed at same temperature and different strain rates. Microstructure observation of as deformed alloy indicated hydrogen promoted DRX on both a and b phase, and encouraged the decomposition of residual lamella. Electron back-scattered diffraction results indicated that hydrogen mainly encouraged discontinues DRX and decreased the dislocation density in a phase. Compared to unhydrogenated alloy, when hydrogen content was 5.31 102 wt.%, volume fraction of DRX increased from 42% to 53% and 41.8%e72.1% at strain rate of 0.01 and 0.001 s1, respectively. © 2017 Elsevier B.V. All rights reserved.
Keywords: Ti-6Al-4V alloy Melt hydrogenation Hot deformation Dynamic recrystallization Dislocation
1. Introduction Ti-6Al-4V alloy (Ti64) with high strength, low density and high corrosion resistance, has been widely used in the industry and/or science of aerospace and aviation [1,2]. It has been reported that over the half titanium products [3] are Ti64 alloys, however, due to the poor workability, manufacturing the Ti64 alloy by cold working induces serious cracks which decides the Ti64 alloy must be hot worked before used. Titanium alloy including Ti64 alloy with high deforming resistance which means improving the hot workability of Ti64 alloy is meaningful and important. Hydrogenation is an effective method to improve the hot workability of titanium alloys. It has been reported [4] that hydrogen as temporary alloying element can increase the volume fraction of softer b phase and dynamic recrystallization (DRX), and then decrease the flow stress of titanium alloy deformed at high temperature. Senkov [5] investigated the thermo-hydrogen
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (L. Wang), [email protected] (Y. Su). http://dx.doi.org/10.1016/j.jallcom.2017.09.044 0925-8388/© 2017 Elsevier B.V. All rights reserved.
processing of titanium alloys, and found that hydrogen enhanced the interaction between dislocation and obstacles, and then increase the dislocations mobility which was beneficial to the high temperature deformation. Chen [6] and Ma [7] hydrogenated the titanium aluminides alloys and found that hydrogen decreased the flow stress and encouraged the discontinuous dynamic recrystallization (DDRX). Most researchers [8e10] completed the hydrogenation process by holding the materials in hydrogen included environment at high temperature (750 C) for several hours. Its low efficiency and longtime cost restricted the wide use. In this study, hydrogen was added into the materials by melting the alloys in gas mixture of hydrogen and argon, hydrogenation and fabrication were completed at the same time. In others work [11e14], this method was called melt hydrogenation, Wang [15,16] investigated the microstructure evaluation, hot deformation behavior and hydride formation in Ti64 alloy. Liu [17e20] studied hydrogen absorption, hot compression and mold filling behavior of TiAl alloy fabricated by melt hydrogenation. The acquired experimental results indicated that hydrogen decreased the flow stress and improved the hot workability of hydrogenated alloys, which was induced by improvement of softer b phase, DRX and dislocation mobility. However, the effect of melt
710
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
hydrogenation on high temperature deformation of Ti64 alloy has not been systematically studied. Therefore, this study focused on the effect of deforming temperature, strain rate and hydrogen content on hot compressing behavior of melt hydrogenated Ti64 alloy, besides, the mechanism of evaluation on DRX and dislocations were also discussed.
TEM observation was conducted on FEI Talos F200X transmission electron microscope, bright field images on observation of dislocations were taken under the same diffraction condition of a tilt is 0 and b tilt is 0 , and EBSD observation was conducted on FEI Quanta 200F scanning electron microscope.
2. Materials and methods
3. Results and discussion
Ti64 alloy was synthesized by vacuum arc melting which was performed on vacuum arc furnace with tungsten gun and watercooled copper plate. Raw materials were titanium sponge (purity 99.9%), AlV alloy (58.14 wt.% vanadium) and aluminum (purity 99.999%). The samples were melted in gas mixture of hydrogen and argon, hydrogen percentage was controlled by JF-2200 system which was able to show the hydrogen partial pressure and total pressure of gas mixture in real time. Based on the results shown on JF-2200 system, hydrogen percentage of 10%, 20% and 30% in gas mixture were obtained. Each specimen was re-melted in gas mixture for 5 times to keep chemical composition homogenous, meanwhile, melting time and input power were kept all the same. Hydrogen content in as received Ti64 alloy was measured by weighting the specimen before and after dehydrogenation process to the accuracy of 0.01 mg, dehydrogenation process was performed on the vacuum annealing furnace at 750 C for 6 h, working pressure was lower than 5 103 Pa. The relative experimental details of dehydrogenation had been published in the previous work from our group [21], verification of hydrogen content and other information can be found elsewhere [16]. This paper will not show the further details, and the relative hydrogen content were 3.64 102 wt.%, 5.31 102 wt.% and 6.39 102 wt.% when hydrogen percentage in gas mixture were 10%, 20% and 30%, respectively. Samples for X-ray diffraction (XRD) analyze were pieces with size of 10 10 2 mm, which were cut from as cast alloys, after mechanical grind, the tests were conducted on D/max-RB X-ray diffractometer with scanning angle ranging from 20 to 80 . The as cast alloys were cut into cylinders with 6 mm in diameter and 9 mm in height, after mechanical grind and cleaned by ultrasonic wave, the cylinder samples were used for high temperature compression experiment. Before hot compression, all the samples were coated with glass lubricant which is a kind of silicone gel with good heat resistance and lubricity at high temperature, to prevent the escape of hydrogen from samples. Hot compression of Ti64 alloy was performed on Gleeble-1500D thermal simulation machine. High purity argon gas was put into the working chamber as protective gas during the hot compression after environmental pressure vacuumized to the lower than 5 104 Pa. The samples were heated from room temperature to evaluated temperature in 10 C/s and then hold at evaluated temperature for 3 min. Hot compressing temperature of 700, 750, 800 and 850 C and strain rate of 0.001, 0.01 and 0.1 s1 were chosen in this study, after hot compression, the samples were quenched into water immediately to preserve the high temperature microstructure. The specimens for microstructure observation were cut from the center of as compressed alloy, after mechanical grind and polished, they were etched by solution of hydrofluoric acid, nitric acid and water (HF:HNO3:H2O is 1:1:8, volume fraction) for 10 s. Scanning electron microscopy (SEM) of as compressed alloy was conducted on FEI Quanta 200F scanning electron microscope in secondary electron mode. Specimens for transmission electron micrograph (TEM) were cut from the center of as compressed alloy, after mechanical grind to 60 mm, they were electrochemical polished by twin-jet polisher in solution of 6% perchloric acid, 34% n-butyl alcohol and 60% methanol (volume friction) at 25 C for 45s, and same solution was used to prepare the samples for electron back-scatter diffraction (EBSD) observation,
3.1. Hot compression at same strain rate Fig. 1 shows the XRD results of as cast alloys with increasing hydrogen content, no hydride was detected due to the low hydrogen content, and the peaks of b phase move to lower angles. Hydrogen is interstitial atom, and the solubility of hydrogen in b phase is much higher than in a phase. In this study, the solutes hydrogen atoms in b phase leading to the lattice expansion which decrease the Bragg angles of b phase. For example, 5.31 102 wt.% hydrogen decreases the Bragg angles of b phase from 39.318 to 38.566 . In comparison, the Bragg angles of a phase keep in the same position. This may be the proofs that hydrogen dissolves into the b phase in as received alloys, as shown in Fig. 1. Fig. 2 shows the true strain-stress curves of unhydrogenated and hydrogenated Ti64 alloy deforming at increasing temperature and same strain rate (0.01 s1). According to the results shown by Zong [22], the a/b transition point of unhydrogenated Ti64 alloy is 980 C. In this study, the highest hydrogen percentage in gas mixture is 30%, and hydrogen content in Ti64 alloy is only 6.39 102 wt.%, therefore, hydrogen can decrease the a/b transition point, but the deforming temperatures (700e850 C) chosen in this paper are still in aþb phase region. According to the results shown in Fig. 2, work hardening effect can be found at the starting stage of compression in which flow stress increases and reaches the peak point. Working hardening is mainly induced by dislocation tangles and piles up. After the peak point, flow stress decreases gradually with increase of strain, which is induced by DRX. The flow stress is sensitive to the change of deforming temperature, flow stress decreases with increase of deforming temperature, which is induced by the increase of softer b phase content and decline of bonding force between atoms. The peak stress of Ti64 alloy with increasing hydrogen content deformed at same strain rate (0.01 s1) and increasing temperature (750 and 850 C) is plotted as Fig. 3. When deformed at higher temperature (850 C), peak stress decreases with increase of
Fig. 1. XRD results of as received Ti64 alloys with increasing hydrogen content. (1) Unhydrogenated; (2) 3.64 102 wt.% hydrogen; (3) 5.31 102 wt.% hydrogen; (4) 6.39 102 wt.% hydrogen.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
711
Fig. 2. True stress-strain curves of Ti64 alloy with increasing hydrogen content deformed at strain rate of 0.01s1 and temperature of 700, 750, 800 and 850 C. (a) Unhydrogenated; (b) 3.64 102 wt.% hydrogen; (c) 5.31 102 wt.% hydrogen; (d) 6.39 102 wt.% hydrogen.
hydrogen content, hydrogen as b phase stabilizing element which can increase the amount of softer b phase, in this paper, the b phase content in as compressed alloys without and with 5.31 102 wt.% hydrogen deformed at 750 and 850 C and 0.01 s1 are calculated from the EBSD data directly, and this paper will not show the further details. Specifically, compared to the alloy without hydrogen, 5.31 102 wt.% hydrogen increases the b phase content from 4.1%
Fig. 3. Peak stress of Ti64 alloy with increasing hydrogen content deformed at strain rate of 0.01 s1 and temperature of 750 and 850 C.
to 4.6% and 4.8% to 5.4 when deformed at 750 and 850 C, respectively. And the b phase with bcc structure has better deformability than a phase with hcp structure, increase on b phase content is beneficial to high temperature deformation. On the other hand, hydrogen can improve the volume fraction of DRX grains which are the main softening way of Ti64 alloy deforming at high temperature. Fig. 4a and b shows the DRX grains of Ti64 alloy with 5.31 102 wt.% hydrogen and deformed at 750 and 850 C, respectively. The original structure of as cast Ti64 alloy is lamella structure which consist of a colony and b lamella, after hot compression, the original a/b interface is broken, and DRX occurs on both a and b phase. As shown in Fig. 4a and b, the DRXed alpha has equixial form and larger grain size due to its high volume fraction, and b phase distributes around the a phase. Due to the low content of b phase, the DRX mainly occurs in a phase. When deformed at 750 C, there are few DRX grains (Fig. 4a) in hydrogenated Ti64 alloy which induces higher peak stress, and when the deforming temperature increases to 850 C, volume fraction of DRX grains increases (Fig. 4b) and peak stress decreases. Higher temperature (850 C) means higher volume fraction of softer b phase and more DRX grains. Therefore, when deformed at higher temperatures (850 C), hydrogen induced softening is the domain effect. When deformed at lower temperatures (750 C), peak stress firstly decreases and then increases with increase of hydrogen content. As discussed earlier, b phase content and volume fraction of DRX grains have close relationship with the deforming temperature, lower deforming temperature decreases the volume fraction
712
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
Fig. 4. DRX grains and dislocations in as compressed Ti64 alloy deformed at: (a) 750 C, 5.31 102 wt.% (b) 850 C, 5.31 102 wt.% (c) 700 C, 5.31 102 wt.%; (d) 700 C, 6.39 102 wt.%.
of b phase and DRX grains, hence the hydrogen still induces softening effect but weaker. When hydrogen content is lower (less than 5.31 102 wt.%), hydrogen induces softening effect, with increase of hydrogen content (6.39 102 wt.%), peak stress increases which is probably induced by hydrogen induced change on dislocations mobility. It has been reported [23] that hydrogen has dragging effect for movement of dislocations, under this circumstance, higher hydrogen (6.39 102 wt.%) content isn't beneficial to the deformation behavior of Ti64 alloy at lower temperature. Besides, lower temperature decreases the dislocation mobility, and there will be more dislocation tangles and piles up in Ti64 alloy with higher hydrogen content (6.39 102 wt.%). Fig. 4c and d shows the dislocations in a phase of Ti64 alloy deformed at 700 C. When hydrogen content is 5.31 102 wt.%, there are few dislocations and no dislocation tangles and piles up are found, as consequence, peak stress decreases. When hydrogen content increases to 6.39 102 wt.%, as shown in Fig. 4d, dislocation tangles and piles up can be seen clearly, which are not beneficial to the high temperature deformation. Therefore, at lower temperature (750 C), proper hydrogen content decreases the peak stress of hot deformed Ti64 alloy. In brief, hydrogen induces softening effect when deforming at higher (850 C) temperature, which is mainly caused by hydrogen induced improvement of b phase content and DRX. When deformed at lower temperature (750 C), there is a competition between hydrogen induced softening and hardening effect. When hydrogen content is lower than 5.31 102 wt.%, peak stress firstly decreases because hydrogen induced DRX consumes most of dislocations, and softening is the dominant effect. When hydrogen content is 6.39 102 wt.%, peak stress increases because hydrogen induces more dislocations tangles and piles up.
3.2. Hot compression at same deforming temperature Fig. 5 shows the peak stress of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.001 to 0.1 s1. Peak stress of Ti64 alloy with same hydrogen content increases with increase of strain rate, because higher strain rate means no enough time for DRX grains to grow up. As shown in Fig. 5, hydrogen decreases peak stress of Ti64 alloy at different strain rate, when strain rate is 0.1 s1, peak stress decreased from 291.89 to 268 MPa as hydrogen content increases to
Fig. 5. Peak stress of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.001 to 0.1 s1.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
5.31 102 wt.%. Same results are found when strain rate is lower, 5.31 102 wt.% hydrogen decreases the peak stress from 214.99 to 121.69 MPa to 147.97 and 106.36 MPa when strain rate are 0.01 and 0.001 s1, respectively. When the Ti64 alloy deformed at high temperature, DRX is the main way to balance out the work hardening effect and decreases the flow stress. It has been reported [6,24] that hydrogen promotes the nucleation and growth of DRX grains during high temperature deformation. The effect of hydrogen on DRX will be discussed later. Fig. 6 shows the microstructure of as compressed Ti64 alloy with different hydrogen content and deformed at 850 C and increasing strain rates. Fig. 5a, c, e and b, d, f are the SEM images of as compressed Ti64 alloy without and with 5.31 102 wt.% hydrogen, respectively. When strain rate is 0.1 s1 (Fig. 6a and b), there is no sufficient time for DRX grains to grow up, and most of microstructure is residual lamella. Compared to the unhydrogenated Ti64 alloy, hydrogen encourages the decomposition of original lamella and formation of new grains. When the strain rate decreases to 0.01 s1,
713
in the as compressed Ti64 alloy without hydrogen (Fig. 6c), most of the structure is still residual lamella, however, in the hydrogenated alloy, many newly formed grains (arrows shown in Fig. 6d) can be seen clearly, which are probably DRX grains. As discussed early, during hot deformation, DRX occurs on both a and b phase, however, due to the low volume fraction of b phase, the DRX mainly occurs in a phase, the newly formed grains shown in Fig. 6d can be distinguished as a phase. At strain rate of 0.001 s1, many newly formed grains can be seen in both as compressed Ti64 alloy without and with hydrogen (Fig. 6e and f). Residual lamella still can be found in the unhydrogenated alloy, but in the hydrogenated alloy, no residual lamella was found, and most of the deformed structure is replaced by newly formed grains, at this circumstance, the hydrogenated alloy mainly consists of primary a grains and transformed b phase. According to the grains size and morphology, the equiaxial grains shown in Fig. 6c, d and e are probably DRX grains, but further verification is necessary. In this paper, the microstructure of as compressed alloy will be analyzed by EBSD technology.
Fig. 6. Microstructure of as compressed Ti64 alloy deformed at 850 C and: (a) (b) 0.1s1; (c) (b) 0.01 s1; (e) (f) 0.001 s1; and (a) (c) (e) without hydrogen; (b) (d) (f) 5.31 102 wt.% hydrogen.
714
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
3.3. Microstructural evaluation of as compressed alloy Fig. 7a and b shows the unique grain color figure (UGCF) maps [25] of as compressed Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate of 0.01 s1. According to the results shown in Fig. 7a and b, as compressed Ti64 alloy without hydrogen have more deformed grains and the grain size shown in Fig. 7a is much bigger than that shown in Fig. 7b, besides, compared to the Ti64 alloy with 5.31 102 wt.% hydrogen, there are less fine and equiaxial grains in the unhydrogenated alloy. Different with the results shown in Fig. 7a, in the as compressed Ti64 alloy with 5.31 102 wt.% hydrogen, most of the deformed structure is replaced by the newly formed grains and the grain size is smaller. However, the grain shape and size can't be the only evidence to judge that whether the newly formed grains are the DRX
grains. Therefore, Fig. 7cef shows the grain orientation spread (GOS) maps and phase content maps with rotation angles from which the volume fraction of DRX grains and phase content can be calculated directly. GOS map shows the average difference in orientation between the average value and the all measured orientation. The volume fraction of DRX region can be measured from the GOS map directly because the GOS value [6,25] of DRX grain is lower than that of deformed grain. As shown in Fig. 7c and d, the blue regions represent the DRX region, and the red parts represent the regions with highest deforming degree. Compared to the hydrogenated alloy (Fig. 7d), there are more deformed grains in unhydrogenated alloy. The volume fraction of DRX increases from 42% in unhydrogenated Ti64 alloy to 53% in Ti64 alloy with 5.31 102 wt.% hydrogen. A possible reason to explain the hydrogen assistated DRX is that hydrogen
Fig. 7. UGCF, GOS and phase content maps of Ti64 alloy deformed at 850 C and strain rate of 0.01 s1. (a), (c), (e) without hydrogen; (b), (d), (f) 5.31 102 wt.% hydrogen.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
refines the microstructure of as cast Ti-6Al-4V alloys [16], which implies there are more grain boundaries and a/b interface. After hot compression, there will be more sub-grains, triple junctions and dislocations walls, which can be the original nucleation sites for DRX. Fig. 7e and f shows the phase content of a and b in deformed Ti64 alloy, the black regions are detected as b phase and the rest parts are a phase, b phase content increases from 4.6% to 6.7% as hydrogen content increases to 5.31 102 wt.%. Compare the GOS maps (Fig. 7c and d) with the phase content maps, DRX (blue regions) mainly occurs in a phase, and b phase is shown as separated particle distributing around the DRXed a phase. Hydrogen addition encourages the DRX and decreases the peak stress of Ti64 alloy
715
when deformed at high temperature. Slight increase on volume fraction of softer b phase may not be sufficient to explain the hydrogen induced softening, therefore, when deformed at 0.01 s1, the drop on flowing stress is caused by hydrogen induced increase on DRX and volume fraction of b phase. Meanwhile, the grain boundaries are also shown in the GOS maps. High angle boundaries (HABs) mainly distribute in the DRX regions (blue parts), and most of the low angle boundaries (LABs) distribute in the deformed regions, other LABs can be found in the DRX regions. That means the DRX grains are also suffering deformation during the hot working of Ti64 alloy. Fig. 8 shows the UGCF and GOS maps of Ti64 alloy without and
Fig. 8. UGCF, GOS and phase content maps of Ti64 alloy deformed at 850 C and strain rate of 0.001 s1. (a), (c), (e) without hydrogen; (b), (d), (f) 5.31 102 wt.% hydrogen.
716
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate of 0.001 s1. When strain rate is 0.001 s1, there is sufficient time for DRX grains to grow up. Both the microstructure of unhydrogenated and hydrogenated alloy are replaced by the fine and equiaxial grains. And the volume fraction of DRX increases from 41.8% in unhydrogenated alloy to 72.1% in hydrogenated alloy. Fig. 8e and f shows the phase content of a and b in Ti64 alloy deformed at 850 C and 0.001 s1, same as the alloy deformed at 0.01 s1, DRX mainly occurs in a phase, and b phase content slightly increases from 4.8% to 5.4%, in comparison, volume fraction of DRX significantly increases, therefore, when deformed at 0.001 s1 and 850 C, softening is mainly caused by hydrogen induced
improvement on DRX in a phase. Meanwhile, most of the HABs distribute in the DRX regions and LABs distribute in the deformed regions. In our previous work [21], we have reported that hydrogen encourages the DRX, and discussed the difference between Discontinuous and Continuous DRX [26e29] in titanium matrix composites. DRX can be classified into two modes, discontinuous dynamic recrystallization (DDRX) and continuous dynamic recrystallization (CDRX). Their difference is distinguished by the formation mode of HABs. DDRX can be found in materials with low stacking fault energy, existing HABs migrate to the side with higher strain energy, during this process, strain energy decreases and dislocations are consumed
Fig. 9. KAM maps of Ti64 alloy deformed at 850 C and strain rate of 0.01 and 0.001 s1. (a) Unhydrogenated, 0.01 s1; (b) 5.31 102 wt.% hydrogen, 0.01 s1; (c) Unhydrogenated, 0.001 s1; (d) 5.31 102 wt.% hydrogen, 0.001 s1;.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
by the newly formed grains. As a result, work hardening is balanced out and flow stress decreases. Unlike the DDRX, the CDRX occurs in materials with higher stacking fault energy, HABs of CDRX are acquired by the transformation from LABs to HABs. Hydrogen can increases the diffusion coefficient of element and migration of grain boundaries in titanium alloys when deformed at high temperature [30], based on this point, hydrogen encourages the DDRX in Ti64 alloy during hot working. Both DDRX and CDRX induce decrease of dislocations density, when deformed at higher strain rate (0.01 s1), there is no enough time for DRX grains to consume the dislocations, as results, there are still plenty of dislocations in DRX regions. Unlike the results shown in Fig. 7c and d, no LABs are found in the DRX regions when strain rate decreases to 0.001 s1. Low strain rate induces the sufficient DRX which means the low density of dislocations. In the other words, the LABs found in DRX regions shown in Fig. 7c and d are probably the stacking of dislocations. To further investigate the change of dislocations density in as compressed Ti64 alloy, Fig. 9 shows the kernel average misorientation (KAM) maps of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.01 to 0.001 s1. The blue and red parts shown in Fig. 9 represent the regions with lowest and highest density of dislocations, respectively. When deformed at higher strain rate (0.01 s1), unhydrogenated alloy has higher density of dislocations, hydrogen addition decreases the dislocations density, and the percentage of regions with low density of dislocations increases from 57.3% in unhydrogenated alloy to 71% in alloy with 5.31 102 wt.% hydrogen. By comparison between Fig. 8a and b with Fig. 7e and f, the regions with high density of dislocations mainly distribute on a phase. Under the same experimental condition, there are more dislocations in a phase than that in b phase, because b phase with bcc structure has better deformability than a phase with hcp structure. Meanwhile, growth of DRX grains consumes most of dislocations, therefore, when deformed at 0.01 s1, decrease on dislocations density is caused by hydrogen induced increase on b phase content and DRX grains. When strain rate is 0.001 s1, as shown in Fig. 9c and d, most parts are regions with low density of dislocations. The percentage of regions with low density of dislocations in unhydrogenated alloy is 84.1%, and it's almost the same as that in hydrogenated alloy (83.3%). Low strain rate induces sufficient DRX which consumes most of dislocations, meanwhile, b phase content with better deformability slightly increases. Therefore, decrease on dislocations density is mainly caused by hydrogen induced increase on DRX, which is also corresponding to decrease of peak stress when strain rate is 0.001 s1. 4. Conclusions In this paper, Ti-6Al-4V alloy was hydrogenated by melt hydrogenation technology, the effect of deforming temperature, strain rate and hydrogen content on hot deformation behavior was investigated. Based on the experimental results and discussion, following conclusions can be obtained. (1) At same strain rate, when deformed at 800 and 850 C, peak stress decreased with increase of hydrogen content, when deformed at 700 and 750 C, peak stress firstly decreased and then increased with increase of hydrogen content. (2) Hydrogen decreased the peak stress and improved hot workability of Ti64 alloy when deformed at same deforming temperature and different strain rate. Hydrogen encouraged the dynamic recrystallization and decomposition of residual lamella.
717
(3) Hydrogen promoted the discontinuous dynamic recrystallization and decreased the dislocations density. When hydrogen content was 5.31 102 wt.%, volume fraction of DRX increased from 42% to 53% and 41.8%e72.1% when deformed at strain rate of 0.01 and 0.001 s1, respectively. Acknowledgement This work was supported by Natural Science Foundation of China 51425402 and 51501048. References [1] F. Yang, B. Gabbitas, Feasibility of producing Ti-6Al-4V alloy for engineering application by powder compact extrusion of blended elemental powder mixtures, J. Alloy Compd. 695 (2017) 1455e1461. [2] F. Weng, H.J. Yu, C.Z. Chen, J.L. Liu, L.J. Zhao, J.J. Dai, Z.H. Zhao, Effect of process parameters on the microstructure evolution and wear property of the laser cladding coatings on Ti-6Al-4V alloy, J. Alloy Compd. 692 (2017) 989e996. [3] L.S. Luo, Y.Q. Su, J.J. Guo, H.Z. Fu, Formation of titanium hydride in Ti-6Al-4V alloy, J. Alloy Compd. 425 (2006) 140e144. [4] J.I. Qazi, O.N. Senkov, J. Rahim, F.H. Froes, Kinetics of martensite decomposition in Ti-6Al-4V-xH alloys, Mater. Sci. Eng. A 359 (2003) 137e149. [5] O.N. Senkov, Thermohydrogen processing of titanium alloys, Int. J. Hydrogen Energy 24 (1999) 565e576. [6] R. Chen, T. Ma, J. Guo, H. Ding, Y. Su, H. Fu, Deformation behavior and microstructural evolution of hydrogenated Ti44Al6Nb alloy during thermocompression at 1373e1523K, Mater. Design 108 (2016) 259e268. [7] T. Ma, R. Chen, D. Zheng, J. Guo, H. Ding, Y. Su, H. Fu, Influence of hightemperature hydrogen charging on microstructure and hot deformability of binary TiAl alloys, J. Alloy Compd. 701 (2017) 399e407. [8] B.G. Yuan, Y.B. Zheng, L.Q. Gong, Q. Chen, M. Lv, Effect of hydrogen content on microstructures and room-temperature compressive properties of TC21 alloy, Mater. Design 94 (2016) 330e337. [9] J.W. Zhao, H. Ding, H.L. Hou, Z.Q. Li, Influence of hydrogen content on hot deformation behavior and microstructural evolution of Ti600 alloy, J. Alloy Compd. 491 (2010) 673e678. [10] D.B. Shan, Y.Y. Zong, T.F. Lu, Y. Lv, Microstructural evolution and formation mechanism of FCC titanium hydride in Ti-6Al-4V-xH alloys, J. Alloy Compd. 427 (2007) 229e234. [11] F.Y. Dong, Y.Q. Su, L.S. Luo, L. Wang, S.J. Wang, L.J. Guo, H.Z. Fu, Enhanced plasticity in Zr-based bulk metallic glasses by hydrogen, Int. J. Hydrogen Energy 37 (2012) 14697e14701. [12] F. Dong, Y. Su, L. Luo, L. Wang, S. Wang, J. Guo, H. Fu, Deoxidation of bulk metallic glasses by hydrogen arc melting, Mater. Lett. 83 (2012) 1e3. [13] F. Dong, S. Lu, Y. Zhang, L. Luo, Y. Su, B. Wang, H. Huang, Q. Xiang, X. Yuan, X. Zuo, Effect of hydrogen addition on the mechanical properties of a bulk metallic glass, J. Alloy Compd. 695 (2017) 3183e3190. [14] Y.Q. Su, S.J. Wang, L.S. Luo, L. Wang, F.Y. Dong, J.J. Guo, H.Z. Fu, Gradient microstructure of TC21 alloy induced by hydrogen during hydrogenation, Int. J. Hydrogen Energy 37 (2012) 19210e19218. [15] Y.Q. Su, L. Wang, L.S. Luo, X.H. Jiang, J.J. Guo, H.Z. Fu, Deoxidation of Titanium alloy using hydrogen, Int. J. Hydrogen Energy 34 (2009) 8958e8963. [16] Y.Q. Su, L.A. Wang, L.S. Luo, X.W. Liu, J.J. Guo, H.Z. Fu, Investigation of melt hydrogenation on the microstructure and deformation behavior of Ti-6Al-4V alloy, Int. J. Hydrogen Energy 36 (2011) 1027e1036. [17] X.W. Liu, Z.L. Zhang, H. Du, F.C. Liu, Z.T. Fan, Effect of hydrogen on mold filling behavior of TiAl alloy, Intermetallics 55 (2014) 102e107. [18] X.W. Liu, F.C. Liu, H. Du, Z.L. Zhang, Q.Z. Cai, Z.T. Fan, Effect of hydrogen on interfacial reaction between TiAl alloy melt and graphite mold, Intermetallics 53 (2014) 107e111. [19] X.W. Liu, Y.Q. Su, L.S. Luo, J.P. Liu, J.J. Guo, H.Z. Fu, Effect of hydrogen on hot deformation behaviors of TiAl alloys, Int. J. Hydrogen Energy 35 (2010) 13322e13328. [20] X.W. Liu, Y.Q. Su, L.S. Luo, K. Li, F.Y. Dong, J.J. Guo, H.Z. Fu, Effect of hydrogen treatment on solidification structures and mechanical properties of TiAl alloys, Int. J. Hydrogen Energy 36 (2011) 3260e3267. [21] X. Wang, L. Wang, L. Luo, H. Yan, X. Li, R. Chen, Y. Su, J. Guo, H. Fu, High temperature deformation behavior of melt hydrogenated (TiBþTiC)/Ti-6Al-4V composites, Mater. Design 121 (2017) 335e344. [22] Y.Y. Zong, D.B. Shan, Y. Lu, B. Guo, Effect of 0.3 wt%H addition on the high temperature deformation behaviors of Ti-6Al-4V alloy, Int. J. Hydrogen Energy 32 (2007) 3936e3940. [23] Y.Z. Chen, X.Y. Ma, X.H. Shi, T. Suo, C. Borchers, K.H. Zhang, F. Liu, R. Kirchheim, Hardening effects in plastically deformed Pd with the addition of H, Scr. Mater. 98 (2015) 48e51. [24] R. Chen, T. Ma, Z. Sun, J. Guo, H. Ding, Y. Su, H. Fu, The hydrogen absorption behavior of high Nb contained titanium aluminides under high pressure and temperature, Int. J. Hydrogen Energy 41 (2016) 13254e13260. [25] S. Dong, R. Chen, J. Guo, H. Ding, Y. Su, H. Fu, Deformation behavior and microstructural evolution of directionally solidified TiAlNb-based alloy during
718
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
thermo-compression at 1373e1573K, Mater. Design 84 (2015) 118e132. [26] R.D. Doherty, D.A. Hughesb, F.J. Humphreysc, J.J. Jonasd, D. Juul Jensene, M.E. Kassnerf, W.E. Kingg, T.R. McNelleyh, H.J. McQueeni, A.D. Rollettj, Current issues in recrystallization: a review, Mater. Sci. Eng. A 238 (1997) 219e274. [27] T. Sakai, A. Belyakov, R. Kaibyshev, H. Miura, J.J. Jonas, Dynamic and postdynamic recrystallization under hot, cold and severe plastic deformation conditions, Prog. Mater. Sci. 60 (2014) 130e207. [28] Y.C. Lin, D.G. He, M.S. Chen, X.M. Chen, C.Y. Zhao, X. Ma, Z.L. Long, EBSD
analysis of evolution of dynamic recrystallization grains and delta phase in a nickel-based superalloy during hot compressive deformation, Mater. Design 97 (2016) 13e24. , A review of dynamic recrystallization phenomena in [29] K. Huang, R.E. Loge metallic materials, Mater. Design 111 (2016) 548e574. [30] X.L. Han, Q. Wang, D.L. Sun, H.X. Zhang, First-principles study of the effect of hydrogen on the Ti self-diffusion characteristics in the alpha Ti-H system, Scr. Mater. 56 (2007) 77e80.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Hot deformation behavior and dynamic recrystallization of melt hydrogenated Ti-6Al-4V alloy Xuan Wang, Liang Wang*, LiangShun Luo, XiaoDong Liu, YingChun Tang, XinZhong Li, RuiRun Chen, YanQing Su**, JingJie Guo, HengZhi Fu National Key Laboratory for Precision Hot Processing of Metals, School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 April 2017 Received in revised form 3 August 2017 Accepted 5 September 2017 Available online 8 September 2017
The effect of hydrogen on hot deformation behavior of Ti-6Al-4V alloy was investigated. Ti-6Al-4V alloy was hydrogenated by melting alloy in gas mixture of hydrogen and argon (melt hydrogenation). Experimental results of hot compression at same strain rate showed hydrogen decreased flow stress at higher deforming temperature, which was attributed to hydrogen induced dislocation movement and dynamic recrystallization (DRX). At lower temperature, peak stress firstly decreased and then increased with increasing hydrogen content. Hydrogen decreased the peak stress and improved the hot workability of alloy deformed at same temperature and different strain rates. Microstructure observation of as deformed alloy indicated hydrogen promoted DRX on both a and b phase, and encouraged the decomposition of residual lamella. Electron back-scattered diffraction results indicated that hydrogen mainly encouraged discontinues DRX and decreased the dislocation density in a phase. Compared to unhydrogenated alloy, when hydrogen content was 5.31 102 wt.%, volume fraction of DRX increased from 42% to 53% and 41.8%e72.1% at strain rate of 0.01 and 0.001 s1, respectively. © 2017 Elsevier B.V. All rights reserved.
Keywords: Ti-6Al-4V alloy Melt hydrogenation Hot deformation Dynamic recrystallization Dislocation
1. Introduction Ti-6Al-4V alloy (Ti64) with high strength, low density and high corrosion resistance, has been widely used in the industry and/or science of aerospace and aviation [1,2]. It has been reported that over the half titanium products [3] are Ti64 alloys, however, due to the poor workability, manufacturing the Ti64 alloy by cold working induces serious cracks which decides the Ti64 alloy must be hot worked before used. Titanium alloy including Ti64 alloy with high deforming resistance which means improving the hot workability of Ti64 alloy is meaningful and important. Hydrogenation is an effective method to improve the hot workability of titanium alloys. It has been reported [4] that hydrogen as temporary alloying element can increase the volume fraction of softer b phase and dynamic recrystallization (DRX), and then decrease the flow stress of titanium alloy deformed at high temperature. Senkov [5] investigated the thermo-hydrogen
* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (L. Wang), [email protected] (Y. Su). http://dx.doi.org/10.1016/j.jallcom.2017.09.044 0925-8388/© 2017 Elsevier B.V. All rights reserved.
processing of titanium alloys, and found that hydrogen enhanced the interaction between dislocation and obstacles, and then increase the dislocations mobility which was beneficial to the high temperature deformation. Chen [6] and Ma [7] hydrogenated the titanium aluminides alloys and found that hydrogen decreased the flow stress and encouraged the discontinuous dynamic recrystallization (DDRX). Most researchers [8e10] completed the hydrogenation process by holding the materials in hydrogen included environment at high temperature (750 C) for several hours. Its low efficiency and longtime cost restricted the wide use. In this study, hydrogen was added into the materials by melting the alloys in gas mixture of hydrogen and argon, hydrogenation and fabrication were completed at the same time. In others work [11e14], this method was called melt hydrogenation, Wang [15,16] investigated the microstructure evaluation, hot deformation behavior and hydride formation in Ti64 alloy. Liu [17e20] studied hydrogen absorption, hot compression and mold filling behavior of TiAl alloy fabricated by melt hydrogenation. The acquired experimental results indicated that hydrogen decreased the flow stress and improved the hot workability of hydrogenated alloys, which was induced by improvement of softer b phase, DRX and dislocation mobility. However, the effect of melt
710
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
hydrogenation on high temperature deformation of Ti64 alloy has not been systematically studied. Therefore, this study focused on the effect of deforming temperature, strain rate and hydrogen content on hot compressing behavior of melt hydrogenated Ti64 alloy, besides, the mechanism of evaluation on DRX and dislocations were also discussed.
TEM observation was conducted on FEI Talos F200X transmission electron microscope, bright field images on observation of dislocations were taken under the same diffraction condition of a tilt is 0 and b tilt is 0 , and EBSD observation was conducted on FEI Quanta 200F scanning electron microscope.
2. Materials and methods
3. Results and discussion
Ti64 alloy was synthesized by vacuum arc melting which was performed on vacuum arc furnace with tungsten gun and watercooled copper plate. Raw materials were titanium sponge (purity 99.9%), AlV alloy (58.14 wt.% vanadium) and aluminum (purity 99.999%). The samples were melted in gas mixture of hydrogen and argon, hydrogen percentage was controlled by JF-2200 system which was able to show the hydrogen partial pressure and total pressure of gas mixture in real time. Based on the results shown on JF-2200 system, hydrogen percentage of 10%, 20% and 30% in gas mixture were obtained. Each specimen was re-melted in gas mixture for 5 times to keep chemical composition homogenous, meanwhile, melting time and input power were kept all the same. Hydrogen content in as received Ti64 alloy was measured by weighting the specimen before and after dehydrogenation process to the accuracy of 0.01 mg, dehydrogenation process was performed on the vacuum annealing furnace at 750 C for 6 h, working pressure was lower than 5 103 Pa. The relative experimental details of dehydrogenation had been published in the previous work from our group [21], verification of hydrogen content and other information can be found elsewhere [16]. This paper will not show the further details, and the relative hydrogen content were 3.64 102 wt.%, 5.31 102 wt.% and 6.39 102 wt.% when hydrogen percentage in gas mixture were 10%, 20% and 30%, respectively. Samples for X-ray diffraction (XRD) analyze were pieces with size of 10 10 2 mm, which were cut from as cast alloys, after mechanical grind, the tests were conducted on D/max-RB X-ray diffractometer with scanning angle ranging from 20 to 80 . The as cast alloys were cut into cylinders with 6 mm in diameter and 9 mm in height, after mechanical grind and cleaned by ultrasonic wave, the cylinder samples were used for high temperature compression experiment. Before hot compression, all the samples were coated with glass lubricant which is a kind of silicone gel with good heat resistance and lubricity at high temperature, to prevent the escape of hydrogen from samples. Hot compression of Ti64 alloy was performed on Gleeble-1500D thermal simulation machine. High purity argon gas was put into the working chamber as protective gas during the hot compression after environmental pressure vacuumized to the lower than 5 104 Pa. The samples were heated from room temperature to evaluated temperature in 10 C/s and then hold at evaluated temperature for 3 min. Hot compressing temperature of 700, 750, 800 and 850 C and strain rate of 0.001, 0.01 and 0.1 s1 were chosen in this study, after hot compression, the samples were quenched into water immediately to preserve the high temperature microstructure. The specimens for microstructure observation were cut from the center of as compressed alloy, after mechanical grind and polished, they were etched by solution of hydrofluoric acid, nitric acid and water (HF:HNO3:H2O is 1:1:8, volume fraction) for 10 s. Scanning electron microscopy (SEM) of as compressed alloy was conducted on FEI Quanta 200F scanning electron microscope in secondary electron mode. Specimens for transmission electron micrograph (TEM) were cut from the center of as compressed alloy, after mechanical grind to 60 mm, they were electrochemical polished by twin-jet polisher in solution of 6% perchloric acid, 34% n-butyl alcohol and 60% methanol (volume friction) at 25 C for 45s, and same solution was used to prepare the samples for electron back-scatter diffraction (EBSD) observation,
3.1. Hot compression at same strain rate Fig. 1 shows the XRD results of as cast alloys with increasing hydrogen content, no hydride was detected due to the low hydrogen content, and the peaks of b phase move to lower angles. Hydrogen is interstitial atom, and the solubility of hydrogen in b phase is much higher than in a phase. In this study, the solutes hydrogen atoms in b phase leading to the lattice expansion which decrease the Bragg angles of b phase. For example, 5.31 102 wt.% hydrogen decreases the Bragg angles of b phase from 39.318 to 38.566 . In comparison, the Bragg angles of a phase keep in the same position. This may be the proofs that hydrogen dissolves into the b phase in as received alloys, as shown in Fig. 1. Fig. 2 shows the true strain-stress curves of unhydrogenated and hydrogenated Ti64 alloy deforming at increasing temperature and same strain rate (0.01 s1). According to the results shown by Zong [22], the a/b transition point of unhydrogenated Ti64 alloy is 980 C. In this study, the highest hydrogen percentage in gas mixture is 30%, and hydrogen content in Ti64 alloy is only 6.39 102 wt.%, therefore, hydrogen can decrease the a/b transition point, but the deforming temperatures (700e850 C) chosen in this paper are still in aþb phase region. According to the results shown in Fig. 2, work hardening effect can be found at the starting stage of compression in which flow stress increases and reaches the peak point. Working hardening is mainly induced by dislocation tangles and piles up. After the peak point, flow stress decreases gradually with increase of strain, which is induced by DRX. The flow stress is sensitive to the change of deforming temperature, flow stress decreases with increase of deforming temperature, which is induced by the increase of softer b phase content and decline of bonding force between atoms. The peak stress of Ti64 alloy with increasing hydrogen content deformed at same strain rate (0.01 s1) and increasing temperature (750 and 850 C) is plotted as Fig. 3. When deformed at higher temperature (850 C), peak stress decreases with increase of
Fig. 1. XRD results of as received Ti64 alloys with increasing hydrogen content. (1) Unhydrogenated; (2) 3.64 102 wt.% hydrogen; (3) 5.31 102 wt.% hydrogen; (4) 6.39 102 wt.% hydrogen.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
711
Fig. 2. True stress-strain curves of Ti64 alloy with increasing hydrogen content deformed at strain rate of 0.01s1 and temperature of 700, 750, 800 and 850 C. (a) Unhydrogenated; (b) 3.64 102 wt.% hydrogen; (c) 5.31 102 wt.% hydrogen; (d) 6.39 102 wt.% hydrogen.
hydrogen content, hydrogen as b phase stabilizing element which can increase the amount of softer b phase, in this paper, the b phase content in as compressed alloys without and with 5.31 102 wt.% hydrogen deformed at 750 and 850 C and 0.01 s1 are calculated from the EBSD data directly, and this paper will not show the further details. Specifically, compared to the alloy without hydrogen, 5.31 102 wt.% hydrogen increases the b phase content from 4.1%
Fig. 3. Peak stress of Ti64 alloy with increasing hydrogen content deformed at strain rate of 0.01 s1 and temperature of 750 and 850 C.
to 4.6% and 4.8% to 5.4 when deformed at 750 and 850 C, respectively. And the b phase with bcc structure has better deformability than a phase with hcp structure, increase on b phase content is beneficial to high temperature deformation. On the other hand, hydrogen can improve the volume fraction of DRX grains which are the main softening way of Ti64 alloy deforming at high temperature. Fig. 4a and b shows the DRX grains of Ti64 alloy with 5.31 102 wt.% hydrogen and deformed at 750 and 850 C, respectively. The original structure of as cast Ti64 alloy is lamella structure which consist of a colony and b lamella, after hot compression, the original a/b interface is broken, and DRX occurs on both a and b phase. As shown in Fig. 4a and b, the DRXed alpha has equixial form and larger grain size due to its high volume fraction, and b phase distributes around the a phase. Due to the low content of b phase, the DRX mainly occurs in a phase. When deformed at 750 C, there are few DRX grains (Fig. 4a) in hydrogenated Ti64 alloy which induces higher peak stress, and when the deforming temperature increases to 850 C, volume fraction of DRX grains increases (Fig. 4b) and peak stress decreases. Higher temperature (850 C) means higher volume fraction of softer b phase and more DRX grains. Therefore, when deformed at higher temperatures (850 C), hydrogen induced softening is the domain effect. When deformed at lower temperatures (750 C), peak stress firstly decreases and then increases with increase of hydrogen content. As discussed earlier, b phase content and volume fraction of DRX grains have close relationship with the deforming temperature, lower deforming temperature decreases the volume fraction
712
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
Fig. 4. DRX grains and dislocations in as compressed Ti64 alloy deformed at: (a) 750 C, 5.31 102 wt.% (b) 850 C, 5.31 102 wt.% (c) 700 C, 5.31 102 wt.%; (d) 700 C, 6.39 102 wt.%.
of b phase and DRX grains, hence the hydrogen still induces softening effect but weaker. When hydrogen content is lower (less than 5.31 102 wt.%), hydrogen induces softening effect, with increase of hydrogen content (6.39 102 wt.%), peak stress increases which is probably induced by hydrogen induced change on dislocations mobility. It has been reported [23] that hydrogen has dragging effect for movement of dislocations, under this circumstance, higher hydrogen (6.39 102 wt.%) content isn't beneficial to the deformation behavior of Ti64 alloy at lower temperature. Besides, lower temperature decreases the dislocation mobility, and there will be more dislocation tangles and piles up in Ti64 alloy with higher hydrogen content (6.39 102 wt.%). Fig. 4c and d shows the dislocations in a phase of Ti64 alloy deformed at 700 C. When hydrogen content is 5.31 102 wt.%, there are few dislocations and no dislocation tangles and piles up are found, as consequence, peak stress decreases. When hydrogen content increases to 6.39 102 wt.%, as shown in Fig. 4d, dislocation tangles and piles up can be seen clearly, which are not beneficial to the high temperature deformation. Therefore, at lower temperature (750 C), proper hydrogen content decreases the peak stress of hot deformed Ti64 alloy. In brief, hydrogen induces softening effect when deforming at higher (850 C) temperature, which is mainly caused by hydrogen induced improvement of b phase content and DRX. When deformed at lower temperature (750 C), there is a competition between hydrogen induced softening and hardening effect. When hydrogen content is lower than 5.31 102 wt.%, peak stress firstly decreases because hydrogen induced DRX consumes most of dislocations, and softening is the dominant effect. When hydrogen content is 6.39 102 wt.%, peak stress increases because hydrogen induces more dislocations tangles and piles up.
3.2. Hot compression at same deforming temperature Fig. 5 shows the peak stress of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.001 to 0.1 s1. Peak stress of Ti64 alloy with same hydrogen content increases with increase of strain rate, because higher strain rate means no enough time for DRX grains to grow up. As shown in Fig. 5, hydrogen decreases peak stress of Ti64 alloy at different strain rate, when strain rate is 0.1 s1, peak stress decreased from 291.89 to 268 MPa as hydrogen content increases to
Fig. 5. Peak stress of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.001 to 0.1 s1.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
5.31 102 wt.%. Same results are found when strain rate is lower, 5.31 102 wt.% hydrogen decreases the peak stress from 214.99 to 121.69 MPa to 147.97 and 106.36 MPa when strain rate are 0.01 and 0.001 s1, respectively. When the Ti64 alloy deformed at high temperature, DRX is the main way to balance out the work hardening effect and decreases the flow stress. It has been reported [6,24] that hydrogen promotes the nucleation and growth of DRX grains during high temperature deformation. The effect of hydrogen on DRX will be discussed later. Fig. 6 shows the microstructure of as compressed Ti64 alloy with different hydrogen content and deformed at 850 C and increasing strain rates. Fig. 5a, c, e and b, d, f are the SEM images of as compressed Ti64 alloy without and with 5.31 102 wt.% hydrogen, respectively. When strain rate is 0.1 s1 (Fig. 6a and b), there is no sufficient time for DRX grains to grow up, and most of microstructure is residual lamella. Compared to the unhydrogenated Ti64 alloy, hydrogen encourages the decomposition of original lamella and formation of new grains. When the strain rate decreases to 0.01 s1,
713
in the as compressed Ti64 alloy without hydrogen (Fig. 6c), most of the structure is still residual lamella, however, in the hydrogenated alloy, many newly formed grains (arrows shown in Fig. 6d) can be seen clearly, which are probably DRX grains. As discussed early, during hot deformation, DRX occurs on both a and b phase, however, due to the low volume fraction of b phase, the DRX mainly occurs in a phase, the newly formed grains shown in Fig. 6d can be distinguished as a phase. At strain rate of 0.001 s1, many newly formed grains can be seen in both as compressed Ti64 alloy without and with hydrogen (Fig. 6e and f). Residual lamella still can be found in the unhydrogenated alloy, but in the hydrogenated alloy, no residual lamella was found, and most of the deformed structure is replaced by newly formed grains, at this circumstance, the hydrogenated alloy mainly consists of primary a grains and transformed b phase. According to the grains size and morphology, the equiaxial grains shown in Fig. 6c, d and e are probably DRX grains, but further verification is necessary. In this paper, the microstructure of as compressed alloy will be analyzed by EBSD technology.
Fig. 6. Microstructure of as compressed Ti64 alloy deformed at 850 C and: (a) (b) 0.1s1; (c) (b) 0.01 s1; (e) (f) 0.001 s1; and (a) (c) (e) without hydrogen; (b) (d) (f) 5.31 102 wt.% hydrogen.
714
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
3.3. Microstructural evaluation of as compressed alloy Fig. 7a and b shows the unique grain color figure (UGCF) maps [25] of as compressed Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate of 0.01 s1. According to the results shown in Fig. 7a and b, as compressed Ti64 alloy without hydrogen have more deformed grains and the grain size shown in Fig. 7a is much bigger than that shown in Fig. 7b, besides, compared to the Ti64 alloy with 5.31 102 wt.% hydrogen, there are less fine and equiaxial grains in the unhydrogenated alloy. Different with the results shown in Fig. 7a, in the as compressed Ti64 alloy with 5.31 102 wt.% hydrogen, most of the deformed structure is replaced by the newly formed grains and the grain size is smaller. However, the grain shape and size can't be the only evidence to judge that whether the newly formed grains are the DRX
grains. Therefore, Fig. 7cef shows the grain orientation spread (GOS) maps and phase content maps with rotation angles from which the volume fraction of DRX grains and phase content can be calculated directly. GOS map shows the average difference in orientation between the average value and the all measured orientation. The volume fraction of DRX region can be measured from the GOS map directly because the GOS value [6,25] of DRX grain is lower than that of deformed grain. As shown in Fig. 7c and d, the blue regions represent the DRX region, and the red parts represent the regions with highest deforming degree. Compared to the hydrogenated alloy (Fig. 7d), there are more deformed grains in unhydrogenated alloy. The volume fraction of DRX increases from 42% in unhydrogenated Ti64 alloy to 53% in Ti64 alloy with 5.31 102 wt.% hydrogen. A possible reason to explain the hydrogen assistated DRX is that hydrogen
Fig. 7. UGCF, GOS and phase content maps of Ti64 alloy deformed at 850 C and strain rate of 0.01 s1. (a), (c), (e) without hydrogen; (b), (d), (f) 5.31 102 wt.% hydrogen.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
refines the microstructure of as cast Ti-6Al-4V alloys [16], which implies there are more grain boundaries and a/b interface. After hot compression, there will be more sub-grains, triple junctions and dislocations walls, which can be the original nucleation sites for DRX. Fig. 7e and f shows the phase content of a and b in deformed Ti64 alloy, the black regions are detected as b phase and the rest parts are a phase, b phase content increases from 4.6% to 6.7% as hydrogen content increases to 5.31 102 wt.%. Compare the GOS maps (Fig. 7c and d) with the phase content maps, DRX (blue regions) mainly occurs in a phase, and b phase is shown as separated particle distributing around the DRXed a phase. Hydrogen addition encourages the DRX and decreases the peak stress of Ti64 alloy
715
when deformed at high temperature. Slight increase on volume fraction of softer b phase may not be sufficient to explain the hydrogen induced softening, therefore, when deformed at 0.01 s1, the drop on flowing stress is caused by hydrogen induced increase on DRX and volume fraction of b phase. Meanwhile, the grain boundaries are also shown in the GOS maps. High angle boundaries (HABs) mainly distribute in the DRX regions (blue parts), and most of the low angle boundaries (LABs) distribute in the deformed regions, other LABs can be found in the DRX regions. That means the DRX grains are also suffering deformation during the hot working of Ti64 alloy. Fig. 8 shows the UGCF and GOS maps of Ti64 alloy without and
Fig. 8. UGCF, GOS and phase content maps of Ti64 alloy deformed at 850 C and strain rate of 0.001 s1. (a), (c), (e) without hydrogen; (b), (d), (f) 5.31 102 wt.% hydrogen.
716
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate of 0.001 s1. When strain rate is 0.001 s1, there is sufficient time for DRX grains to grow up. Both the microstructure of unhydrogenated and hydrogenated alloy are replaced by the fine and equiaxial grains. And the volume fraction of DRX increases from 41.8% in unhydrogenated alloy to 72.1% in hydrogenated alloy. Fig. 8e and f shows the phase content of a and b in Ti64 alloy deformed at 850 C and 0.001 s1, same as the alloy deformed at 0.01 s1, DRX mainly occurs in a phase, and b phase content slightly increases from 4.8% to 5.4%, in comparison, volume fraction of DRX significantly increases, therefore, when deformed at 0.001 s1 and 850 C, softening is mainly caused by hydrogen induced
improvement on DRX in a phase. Meanwhile, most of the HABs distribute in the DRX regions and LABs distribute in the deformed regions. In our previous work [21], we have reported that hydrogen encourages the DRX, and discussed the difference between Discontinuous and Continuous DRX [26e29] in titanium matrix composites. DRX can be classified into two modes, discontinuous dynamic recrystallization (DDRX) and continuous dynamic recrystallization (CDRX). Their difference is distinguished by the formation mode of HABs. DDRX can be found in materials with low stacking fault energy, existing HABs migrate to the side with higher strain energy, during this process, strain energy decreases and dislocations are consumed
Fig. 9. KAM maps of Ti64 alloy deformed at 850 C and strain rate of 0.01 and 0.001 s1. (a) Unhydrogenated, 0.01 s1; (b) 5.31 102 wt.% hydrogen, 0.01 s1; (c) Unhydrogenated, 0.001 s1; (d) 5.31 102 wt.% hydrogen, 0.001 s1;.
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
by the newly formed grains. As a result, work hardening is balanced out and flow stress decreases. Unlike the DDRX, the CDRX occurs in materials with higher stacking fault energy, HABs of CDRX are acquired by the transformation from LABs to HABs. Hydrogen can increases the diffusion coefficient of element and migration of grain boundaries in titanium alloys when deformed at high temperature [30], based on this point, hydrogen encourages the DDRX in Ti64 alloy during hot working. Both DDRX and CDRX induce decrease of dislocations density, when deformed at higher strain rate (0.01 s1), there is no enough time for DRX grains to consume the dislocations, as results, there are still plenty of dislocations in DRX regions. Unlike the results shown in Fig. 7c and d, no LABs are found in the DRX regions when strain rate decreases to 0.001 s1. Low strain rate induces the sufficient DRX which means the low density of dislocations. In the other words, the LABs found in DRX regions shown in Fig. 7c and d are probably the stacking of dislocations. To further investigate the change of dislocations density in as compressed Ti64 alloy, Fig. 9 shows the kernel average misorientation (KAM) maps of Ti64 alloy without and with 5.31 102 wt.% hydrogen deformed at 850 C and strain rate ranging from 0.01 to 0.001 s1. The blue and red parts shown in Fig. 9 represent the regions with lowest and highest density of dislocations, respectively. When deformed at higher strain rate (0.01 s1), unhydrogenated alloy has higher density of dislocations, hydrogen addition decreases the dislocations density, and the percentage of regions with low density of dislocations increases from 57.3% in unhydrogenated alloy to 71% in alloy with 5.31 102 wt.% hydrogen. By comparison between Fig. 8a and b with Fig. 7e and f, the regions with high density of dislocations mainly distribute on a phase. Under the same experimental condition, there are more dislocations in a phase than that in b phase, because b phase with bcc structure has better deformability than a phase with hcp structure. Meanwhile, growth of DRX grains consumes most of dislocations, therefore, when deformed at 0.01 s1, decrease on dislocations density is caused by hydrogen induced increase on b phase content and DRX grains. When strain rate is 0.001 s1, as shown in Fig. 9c and d, most parts are regions with low density of dislocations. The percentage of regions with low density of dislocations in unhydrogenated alloy is 84.1%, and it's almost the same as that in hydrogenated alloy (83.3%). Low strain rate induces sufficient DRX which consumes most of dislocations, meanwhile, b phase content with better deformability slightly increases. Therefore, decrease on dislocations density is mainly caused by hydrogen induced increase on DRX, which is also corresponding to decrease of peak stress when strain rate is 0.001 s1. 4. Conclusions In this paper, Ti-6Al-4V alloy was hydrogenated by melt hydrogenation technology, the effect of deforming temperature, strain rate and hydrogen content on hot deformation behavior was investigated. Based on the experimental results and discussion, following conclusions can be obtained. (1) At same strain rate, when deformed at 800 and 850 C, peak stress decreased with increase of hydrogen content, when deformed at 700 and 750 C, peak stress firstly decreased and then increased with increase of hydrogen content. (2) Hydrogen decreased the peak stress and improved hot workability of Ti64 alloy when deformed at same deforming temperature and different strain rate. Hydrogen encouraged the dynamic recrystallization and decomposition of residual lamella.
717
(3) Hydrogen promoted the discontinuous dynamic recrystallization and decreased the dislocations density. When hydrogen content was 5.31 102 wt.%, volume fraction of DRX increased from 42% to 53% and 41.8%e72.1% when deformed at strain rate of 0.01 and 0.001 s1, respectively. Acknowledgement This work was supported by Natural Science Foundation of China 51425402 and 51501048. References [1] F. Yang, B. Gabbitas, Feasibility of producing Ti-6Al-4V alloy for engineering application by powder compact extrusion of blended elemental powder mixtures, J. Alloy Compd. 695 (2017) 1455e1461. [2] F. Weng, H.J. Yu, C.Z. Chen, J.L. Liu, L.J. Zhao, J.J. Dai, Z.H. Zhao, Effect of process parameters on the microstructure evolution and wear property of the laser cladding coatings on Ti-6Al-4V alloy, J. Alloy Compd. 692 (2017) 989e996. [3] L.S. Luo, Y.Q. Su, J.J. Guo, H.Z. Fu, Formation of titanium hydride in Ti-6Al-4V alloy, J. Alloy Compd. 425 (2006) 140e144. [4] J.I. Qazi, O.N. Senkov, J. Rahim, F.H. Froes, Kinetics of martensite decomposition in Ti-6Al-4V-xH alloys, Mater. Sci. Eng. A 359 (2003) 137e149. [5] O.N. Senkov, Thermohydrogen processing of titanium alloys, Int. J. Hydrogen Energy 24 (1999) 565e576. [6] R. Chen, T. Ma, J. Guo, H. Ding, Y. Su, H. Fu, Deformation behavior and microstructural evolution of hydrogenated Ti44Al6Nb alloy during thermocompression at 1373e1523K, Mater. Design 108 (2016) 259e268. [7] T. Ma, R. Chen, D. Zheng, J. Guo, H. Ding, Y. Su, H. Fu, Influence of hightemperature hydrogen charging on microstructure and hot deformability of binary TiAl alloys, J. Alloy Compd. 701 (2017) 399e407. [8] B.G. Yuan, Y.B. Zheng, L.Q. Gong, Q. Chen, M. Lv, Effect of hydrogen content on microstructures and room-temperature compressive properties of TC21 alloy, Mater. Design 94 (2016) 330e337. [9] J.W. Zhao, H. Ding, H.L. Hou, Z.Q. Li, Influence of hydrogen content on hot deformation behavior and microstructural evolution of Ti600 alloy, J. Alloy Compd. 491 (2010) 673e678. [10] D.B. Shan, Y.Y. Zong, T.F. Lu, Y. Lv, Microstructural evolution and formation mechanism of FCC titanium hydride in Ti-6Al-4V-xH alloys, J. Alloy Compd. 427 (2007) 229e234. [11] F.Y. Dong, Y.Q. Su, L.S. Luo, L. Wang, S.J. Wang, L.J. Guo, H.Z. Fu, Enhanced plasticity in Zr-based bulk metallic glasses by hydrogen, Int. J. Hydrogen Energy 37 (2012) 14697e14701. [12] F. Dong, Y. Su, L. Luo, L. Wang, S. Wang, J. Guo, H. Fu, Deoxidation of bulk metallic glasses by hydrogen arc melting, Mater. Lett. 83 (2012) 1e3. [13] F. Dong, S. Lu, Y. Zhang, L. Luo, Y. Su, B. Wang, H. Huang, Q. Xiang, X. Yuan, X. Zuo, Effect of hydrogen addition on the mechanical properties of a bulk metallic glass, J. Alloy Compd. 695 (2017) 3183e3190. [14] Y.Q. Su, S.J. Wang, L.S. Luo, L. Wang, F.Y. Dong, J.J. Guo, H.Z. Fu, Gradient microstructure of TC21 alloy induced by hydrogen during hydrogenation, Int. J. Hydrogen Energy 37 (2012) 19210e19218. [15] Y.Q. Su, L. Wang, L.S. Luo, X.H. Jiang, J.J. Guo, H.Z. Fu, Deoxidation of Titanium alloy using hydrogen, Int. J. Hydrogen Energy 34 (2009) 8958e8963. [16] Y.Q. Su, L.A. Wang, L.S. Luo, X.W. Liu, J.J. Guo, H.Z. Fu, Investigation of melt hydrogenation on the microstructure and deformation behavior of Ti-6Al-4V alloy, Int. J. Hydrogen Energy 36 (2011) 1027e1036. [17] X.W. Liu, Z.L. Zhang, H. Du, F.C. Liu, Z.T. Fan, Effect of hydrogen on mold filling behavior of TiAl alloy, Intermetallics 55 (2014) 102e107. [18] X.W. Liu, F.C. Liu, H. Du, Z.L. Zhang, Q.Z. Cai, Z.T. Fan, Effect of hydrogen on interfacial reaction between TiAl alloy melt and graphite mold, Intermetallics 53 (2014) 107e111. [19] X.W. Liu, Y.Q. Su, L.S. Luo, J.P. Liu, J.J. Guo, H.Z. Fu, Effect of hydrogen on hot deformation behaviors of TiAl alloys, Int. J. Hydrogen Energy 35 (2010) 13322e13328. [20] X.W. Liu, Y.Q. Su, L.S. Luo, K. Li, F.Y. Dong, J.J. Guo, H.Z. Fu, Effect of hydrogen treatment on solidification structures and mechanical properties of TiAl alloys, Int. J. Hydrogen Energy 36 (2011) 3260e3267. [21] X. Wang, L. Wang, L. Luo, H. Yan, X. Li, R. Chen, Y. Su, J. Guo, H. Fu, High temperature deformation behavior of melt hydrogenated (TiBþTiC)/Ti-6Al-4V composites, Mater. Design 121 (2017) 335e344. [22] Y.Y. Zong, D.B. Shan, Y. Lu, B. Guo, Effect of 0.3 wt%H addition on the high temperature deformation behaviors of Ti-6Al-4V alloy, Int. J. Hydrogen Energy 32 (2007) 3936e3940. [23] Y.Z. Chen, X.Y. Ma, X.H. Shi, T. Suo, C. Borchers, K.H. Zhang, F. Liu, R. Kirchheim, Hardening effects in plastically deformed Pd with the addition of H, Scr. Mater. 98 (2015) 48e51. [24] R. Chen, T. Ma, Z. Sun, J. Guo, H. Ding, Y. Su, H. Fu, The hydrogen absorption behavior of high Nb contained titanium aluminides under high pressure and temperature, Int. J. Hydrogen Energy 41 (2016) 13254e13260. [25] S. Dong, R. Chen, J. Guo, H. Ding, Y. Su, H. Fu, Deformation behavior and microstructural evolution of directionally solidified TiAlNb-based alloy during
718
X. Wang et al. / Journal of Alloys and Compounds 728 (2017) 709e718
thermo-compression at 1373e1573K, Mater. Design 84 (2015) 118e132. [26] R.D. Doherty, D.A. Hughesb, F.J. Humphreysc, J.J. Jonasd, D. Juul Jensene, M.E. Kassnerf, W.E. Kingg, T.R. McNelleyh, H.J. McQueeni, A.D. Rollettj, Current issues in recrystallization: a review, Mater. Sci. Eng. A 238 (1997) 219e274. [27] T. Sakai, A. Belyakov, R. Kaibyshev, H. Miura, J.J. Jonas, Dynamic and postdynamic recrystallization under hot, cold and severe plastic deformation conditions, Prog. Mater. Sci. 60 (2014) 130e207. [28] Y.C. Lin, D.G. He, M.S. Chen, X.M. Chen, C.Y. Zhao, X. Ma, Z.L. Long, EBSD
analysis of evolution of dynamic recrystallization grains and delta phase in a nickel-based superalloy during hot compressive deformation, Mater. Design 97 (2016) 13e24. , A review of dynamic recrystallization phenomena in [29] K. Huang, R.E. Loge metallic materials, Mater. Design 111 (2016) 548e574. [30] X.L. Han, Q. Wang, D.L. Sun, H.X. Zhang, First-principles study of the effect of hydrogen on the Ti self-diffusion characteristics in the alpha Ti-H system, Scr. Mater. 56 (2007) 77e80.