This investigation focused on the deformation response and microstructural changes of severely deformed titanium during post-severe plastic deformation tension, at temperatures of 300–600 °C and at strain rates of 0.001–0.1 s−1. The obtained results suggest that SPD enhances the strength of grade 4 titanium up to 500 °C. At above 600 °C, the severely deformed microstructure showed comprehensive recovery. Severely deformed titanium was seen to be highly sensitive to the deformation rate, where strain rate sensitivity increased with the increase of test temperature. Analysis of fracture surfaces reveals that at elevated temperatures, growth of dimples and void coalescence occurs due to the enhanced diffusion rate and occurrence of recrystallized grains.
Due to the superior biocompatibility and relatively low density of titanium (Ti) and its alloys, these materials are very attractive for engineering applications. Lately, substantial efforts have been made to enhance the strength of commercial purity (CP) Ti with the minimum sacrifice of ductility [1–6]. Specifically, severe plastic deformation (SPD) processes were observed to noticeably improve the strength of metallic materials [1–4]. Among SPD processes, the equal channel angular extrusion/pressing (ECAE/P) method was recognized to provide excellent mechanical properties [1,2,7]. The considerable strength enhancement of metallic materials processed via ECAE was a strong motivation for testing the influence of this SPD method on the mechanical properties, microstructure, and texture development of hexagonal close-packed (hcp) metals [1,2,7–9]. Specifically, the monotonic properties and anisotropy of CP titanium after ECAE followed by cold rolling have been investigated . The optimal mechanical properties were achieved in the flow plane perpendicular to the rolling direction. Apart from the enhancement of mechanical properties, the texture development of ECAE processed CP Ti has been analyzed . Accordingly, the development of a strong texture is expected upon the ECAE processing of CP Ti. The texture of CP titanium has been also simulated utilizing viscoplastic self-consistent simulations. Impacts of ECAE routes on mechanical properties and microstructure evolution have also been explored [1,11]. The so-called routes Bc and E were seen to be the most efficient routes in grain refinement upon ECAE providing a considerable fully worked zone in a billet.
Many studies have focused on the stability of severely deformed titanium at high temperatures since this material has the substantial potential of being used for various engineering applications [12–14]. Heat treatment of ultrafine-grained (UFG) titanium in the range of 450–600 °C for 6 h led to the grain growth of the severely deformed titanium . A drop of tensile strength with the rise in test temperature was additionally reported for the UFG counterpart of this metal . It was reported that dislocation recovery caused this continuous decline in the tensile strength. Recently, compression tests have been conducted at a deformation rate of 10 s−1 and the temperatures of 25–400 °C to study the microstructural stabilization and mechanical properties of UFG titanium . The results suggest that recrystallization took place at temperatures above 300 °C. There are also several reports on the warm and hot characteristics of severely deformed titanium [15–18]. Meredith and Khan mentioned that considerable grain growth resulted in a noticeable reduction in strength at deformation temperatures above 500 °C. On a work concerning UFG grade 1 titanium, the compressive behavior was revealed up to 375 °C, both in quasi-static and dynamic regimes . In addition to workability characteristics, microstructure evolution and modeling efforts and damping responses of severely deformed titanium have also been topics of interest [16,17,19–21]. It was stated that deformation of severely deformed titanium at and less than 700 °C leads to flow localization . Moreover, the stress–strain curves of UFG titanium were well-predicted based on the Avrami equations . Modified Johnson–Cook (JC) model as a phenomenological model has also been employed to predict the flow behavior of UFG Ti at elevated temperatures . It was seen that the JC model is capable of predicting the monotonic behavior of UFG Ti with error levels of less than 5%. Not only monotonic behavior but also cyclic deformation behavior of UFG Ti has been studied at elevated temperatures . Results revealed that cyclic stability of severely deformed titanium was deteriorated in comparison to the CG material at temperatures above 400 °C.
Earlier studies on severely deformed hcp materials in general and CP titanium, in particular, demonstrate that the elevated temperature flow behavior is considerably affected by the ECAE processing route and the test temperature and strain rate. Although some research works have concentrated on the elevated temperature behavior of ECAE processed Ti, the lack of a comparative study about the implications of ECAE processing on the elevated temperature flow behavior and the fracture morphology motivated this study.
The as-received CP grade 4 titanium featured a bar shape with the chemical composition shown in Table 1. A graphite-based lubricant was used prior to extrusion. A furnace was employed to heat the bars up to the forming temperature, where they were kept for 60 min prior to ECAE. The lowest extrusion temperature that enabled ECAE processing in a 90 deg ECAE die in the absence of shear localization and macroscopic cracks was found to be 450 °C. The rate of extrusion was measured to be 1.27 mm/s. ECAE process was repeated for eight passes gaining a cumulative strain of 9.24 in the extruded billet . Route E was chosen as the ECAE processing route. It was previously reported that route E can result in a noticeable fully worked zone in a billet . Details of the ECAE route used in the present research work can be found in Ref. . The lowest possible processing temperature was employed to prevent recrystallization in an ECAE die incorporating the notion of sliding walls .
Electro-discharge machining (EDM) was employed to cut test specimens with their tension axis parallel to the extrusion direction. Gauge section dimensions of test specimens were 15 mm × 6 mm × 2 mm. Test specimens were ground and polished to eliminate the effect of the remaining surfaces affected by EDM. The tension experiments were conducted at strain rates of 0.001, 0.01, and 0.1 s−1 and at temperatures of 25, 300, 400, 500, and 600 °C. Tension experiments were carried out inside a temperature-controlled furnace mounted on a servo-hydraulic Instron 8872 mechanical test frame equipped with a load cell (±25 kN). Samples were heated up to the test temperature and were strained up to fracture. Strain values were measured by a hot mountable furnace extensometer. During testing, K-type thermocouples were employed to monitor the specimen temperature. Three tension experiments per condition were conducted to analyze the reproducibility. As the results did not show any pronounced scatter, only one curve for each deformation condition is represented for the sake of clarity.
To capture the microstructure of deformed samples, standard metallographic procedures (ASTM E3) including grinding and polishing were utilized. Specimens were chemically etched by immersing them in Kroll’s reagent. A scanning electron microscope (SEM) was used to image the fracture surfaces and microstructure of the samples in the various states. During microstructural analysis, the acceleration voltage of SEM was set to 12 kV.
3 Results and Discussion
3.1 Mechanical Properties of Severely Deformed Ti at Ambient and Elevated Temperature.
The tensile stress–strain curves of the severely deformed titanium at various temperatures of 25–600 °C and strain rates of 0.001–0.1 s−1 are displayed in Fig. 1. With the increase in test temperature, a significant reduction in true tensile strength values is evident. Such a reduction in flow stress at a higher temperature might be connected to the temperature dependence of diffusion related softening mechanisms, i.e., dynamic recovery (DRV) and dynamic recrystallization (DRX) [19,25–27]. It is well-established that the movement of dislocation, which in turn promotes dynamic softening mechanisms can be accelerated at elevated temperatures [17,28,29]. It should also be noted that at 600 °C, ECAE processing enabled higher elongation at break reaching over 10%. Such a high ductility could be ascribed to the occurrence of DRV and DRX at this deformation temperature. Previously, UFG Ti rods showed higher ductility at elevated temperatures . Changes in grain boundary (GB) structure were found to be the main reason for the observation of high ductility at elevated temperatures .
From Fig. 1, the impact of deformation rate on the flow stress curves of the severely deformed Ti can also be discussed. Accordingly, flow stress levels are more sensitive to the strain rate when the test temperature reaches and exceeds 400 °C. Particularly, a reduction in the rate of deformation causes a drop in the flow stress levels. It is well-documented that lower rates of deformation weaken the inclination for defect interactions and afford a longer time for recovery and recrystallization of the deformed structure [17,28,31]. Another crucial observation that requires to be considered is the variation of the ultimate elongation with the rate of deformation at various temperatures. At all deformation temperatures except 600 °C, ductility of the severely deformed Ti is improved with the decrease of strain rate. At 600 °C, although stress levels are very sensitive to the strain rate, ultimate elongation values are almost identical at various rates of deformation.
To analyze the impact of ECAE processing on the temperature-dependent characteristics of CP Ti, elevated temperature tensile tests were carried out on both as-received and severely deformed CP Ti at a deformation rate of 0.01 s−1 and at temperatures of 300–600 °C (Fig. 2). From Fig. 2, it is clear that ECAE is able to improve the strength of CP Ti at or below 500 °C. The high density of dislocations and GBs was mentioned to be the main reason for the rise in flow strength [32,33]. An investigation on Ti–6Al–4V alloy reported that SPD could enhance the mechanical properties of this alloy up to 500 °C compared with the coarse counterpart; the impact of severely deformed microstructure disappeared at such a high temperature because of the occurrence of recrystallization and subsequent grain growth . The improvement in the strength of severely deformed materials during elevated temperature deformation was also mentioned for the temperature range of 300–450 °C for an aluminum alloy . At 600 °C, the strength of severely deformed samples was seen to be akin to that of the as-received one. This behavior could be imputed to the occurrence of DRX at such an elevated temperature . It is well-established that a high fraction of sub-grain and GBs as well as high dislocation density in UFG metals results in severe recovery and recrystallization while exposing elevated temperatures [14,16]. Furthermore, ductility of severely deformed and as-received counterparts at various deformation temperatures can be compared. SPD was seen to have detrimental impacts on the ductility of CP Ti at temperatures below 600 °C, although both as-received and severely deformed specimens exhibited relatively similar ductility at 400 °C. Embrittlement observed at 500 °C might be linked to the segregation of oxygen and other impurities in GBs [34,35]. The in-depth analysis of the diffusion of oxygen and other impurities along GBs needs to be considered. This is out of the scope of present work and hence will be addressed in future studies. Besides embrittlement induced by impurities, the behavior observed may be attributed to the confinement of damage induced by localized recovery/grain coarsening/recrystallization of the UFG microstructure [36,37]. Considering the theory of damage localization, at a given region in microstructure, plasticity is somehow more pronounced. The increased dislocation density reaches a critical value for the onset of recrystallization in this region. Afterward, strength is locally decreased and eventually leads to damage localization. At the deformation temperature of 600 °C, severely deformed samples exhibited a more ductile response than the CG samples. This observation can be linked to the activation of GB-mediated mechanisms as well as to the progress in dislocation glide with multiple slip systems [32,38,39]. At elevated temperatures, grain boundary diffusion mechanisms such as grain boundary sliding (GBS) should be taken into account. As UFG materials contain a relatively higher density of boundaries, GBS might be the controlling deformation mechanism and thus shorter pathways of diffusion [39,40]. On the other hand, the low ductility of the as-received specimen may be imputed to the hot embrittlement of the coarse-grained structure at high temperatures . It was stated that DRX occurred in the UFG structure to a higher degree, which hinders the nucleation and propagation of micro-voids and cracks.
Peak stress values versus deformation temperature at different deformation rates are shown in Fig. 3(a). Peak stress values remarkably drop with rising deformation temperature at all rates of deformation. This behavior can be rationalized in terms of reduction in dislocation interactions resulting in a lower work hardening (WH) rate . Figure 3(b) displays the deviation of elongation to fracture with temperature. At all rates of deformation, samples deformed at 600 °C shows the highest ductility, attesting the activation of the GBS mechanism at such a high temperature.
Strain rate sensitivity (m) is a key property to investigate the workability and formability of metals at different temperatures . m is defined as the influence of the rate of deformation on dislocation generation and propagation, and in essence, it is an indication of workability. m values are obtained from the slope of the plot at constant temperature and strain. Figure 4 demonstrates the changes in strain rate sensitivity versus strain. As shown in Fig. 4, m value considerably escalates with the rise in temperature as a result of thermally activated mechanisms. It was reported that the higher rate of diffusion related mechanisms and dynamic recrystallization at elevated temperatures lead to higher sensitivity to the rate of deformation [42,43].
3.2 Microstructure Evolution.
To investigate the impact of ECAE processing on the microstructure evolution of CP titanium, SEM analysis was conducted. The SEM images of the CG and UFG CP titanium after tensile tests at various temperatures are demonstrated in Fig. 5. Evidently, the microstructure of the annealed sample is decorated with coarse grains. Obviously, deforming coarse-grained Ti at a higher temperature (600 °C) resulted in a gradual grain growth (cf. Figs. 5(a) and 5(b)). ECAE processing led to a considerable grain refinement as shown in Fig. 5(c). Figures 5(d)–5(h) exhibit the effect of test temperature on the variations in the microstructure of the severely deformed titanium. Results suggest that a slight grain growth occurred at temperatures below 600 °C while a considerable grain growth is evident for the sample deformed at the highest test temperature. The microstructure of the sample tensioned at 600 °C consists of equiaxed grains that can be identified as DRXed structures. The coarsening of dynamically recrystallized grains with the rise in deformation temperature up to 600 °C was previously reported for UFG Ti elsewhere .
The impact of deformation rate on the microstructure evolution of ECAE processed Ti was also analyzed as displayed in Fig. 6. The rate of deformation is not very influential at lower temperatures while conducting tensile tests at 600 °C for longer durations resulted in a coarser microstructure as dynamically recrystallized grains had sufficient time for the growth of newly nucleated grains (Figs. 6(a)–6(d)). Elevated temperatures provide a noticeable driving force for DRX within the microstructure , and thus exposure to high temperature (600 °C) for longer durations caused a coarse microstructure with the average grain size of 9 µm.
3.3 Analysis of Fracture Surfaces.
To understand the fracture behavior of UFG Ti during the elevated temperature tensile tests, SEM studies were also conducted on the fracture surfaces of deformed specimens. Figure 7 illustrates the impact of test temperature on fracture surfaces of severely deformed Ti. All micrographs were taken at the same magnification to make the comparisons more meaningful. The fracture surfaces of all specimens show a remarkable number of micro-voids and equiaxed dimples verifying the occurrence of ductile fracture. At room temperature, dimpled rupture on fracture surfaces is evident, which implies the presence of localized deformation in UFG Ti. Fracture behavior of severely deformed Ti followed by tensile tests at room temperature was earlier investigated elsewhere stating similar findings on fracture morphology of this metal . It should be noted that the size of dimples and micro-voids substantially grows with the rise in deformation temperature. Fracture morphology of the sample deformed at 25 °C consists of a dense dimple structure. When the deformation temperature increases, void coalescence occurs as a fracture mechanism . This can be related to the improvement of the diffusion rate at higher deformation temperatures [47,48]. It is also well-known that the micro-voids coalescence is an internal necking mechanism, which takes place regularly at low-to-moderate stress values in the case of elevated temperature deformation [49,50].
The influence of the strain rate on fracture surfaces of severely deformed Ti was also probed as displayed in Fig. 8. Similar to the case of grain size (Fig. 6), deforming Ti at various deformation rates and at a temperature of 300 °C did not exhibit a different fracture morphology (Figs. 8(a) and 8(b)). However, deforming samples at 600 °C and various strain rates resulted in significant changes in the size of dimples and micro-voids. Accordingly, larger voids and dimples at lower deformation rates are evident. The growth of dimples and voids happens with the reduction in the rate of deformation as a lower strain rate provides enough time for coalescence and growth . Furthermore, higher deformation temperature (600 °C) promotes internal necking ability causing higher resistance of the structure to cavity formation and thus higher ductility . Hence, microstructural analysis on the fracture surfaces supports the higher ductility seen at 600 °C.
Tensile experiment results and the corresponding grain and dimple sizes at different deformation conditions are summarized in Table 2. With a rise in deformation temperature or reduction in strain rate, grain and dimple sizes of severely deformed titanium increase, while its strength is deteriorated. Specifically, at deformation temperatures above 400 °C, peak stress sharply decreased, whereas grain and dimple sizes considerably rose. An inverse power-law relationship between the strength of severely deformed titanium and recrystallized grain size was introduced elsewhere . Accordingly, similar to ambient conditions, an increase in grain size results in a noticeable drop of flow strength for severely deformed titanium at elevated temperatures.
The mechanical properties and fracture behavior of ECAE processed CP titanium were explored at various strain rates and temperatures. The following conclusions can be obtained:
The flow behavior is strongly temperature and deformation rate dependent. The strength of UFG titanium reduces with rising test temperature and drop of deformation rate although the rate of deformation is less influential at ambient temperature.
The strain rate sensitivity (m) of ECAE processed pure titanium was observed to be significantly influenced by the temperature. Strain rate sensitivity remarkably rises with the increase of test temperature up to 600 °C.
Microstructural observations displayed high thermal stability against coarsening up to 500 °C. However, grain growth resulting from grain boundary migration takes place at or above 500 °C. At 600 °C, the thermal stability of severely deformed titanium was observed to be sensitive to the rate of deformation.
Characterization of fracture surfaces implies that ductile fracture occurred at the deformation rate and temperature range employed. Lower rate of deformation and higher deformation temperature resulted in a high tendency of void growth and coalescence.
The authors are thankful for the partial support from the Marie Curie Career Integration Grant (Grant No. 304150) within the EU-FP7 program for conducting this study. The authors are also grateful to the “Safer Materials” project funded by the Hessen State Ministry of Higher Education, Research and the Arts—Initiative for the Development of Scientific and Economic Excellence (LOEWE).
Conflict of Interest
There are no conflicts of interest.