INTRODUCTION

The resistance of materials against fatigue failure is the most important characteristic of the constructive strength of materials. Therefore, numerous publications are devoted to investigating the fatigue strength and fatigue failure mechanism of ultrafine-grained (UFG) materials formed by severe plastic deformation (SPD) [1–14]. Strength increases and ductility decreases after the SPD, which usually leads to an increase in the fatigue limit of materials, but also a decrease or insufficient increase in the fatigue strength in a low-cycle fatigue region [6, 7, 10, 15].

The authors in [7] and some other researchers [16, 17] showed that the UFG materials usually have higher crack propagation velocity in the time-independent mode when compared with coarse-grained (CG) alloy. The authors in [2, 7] associate the low strength of UFG materials in the low-cycle fatigue region with the smaller convolution of the propagating crack, which leads to a decrease in the roughness of cleavage surfaces [7] and with the relatively developed cyclic plastic deformation implemented near the crack vertex [2]. The low plasticity of UFG alloys unfavorably affects the low-cycle fatigue resistance, which strongly limits their practical application. Therefore, the most important direction of increasing the fatigue strength of UFG materials is the development of optimal SPD modes increasing their ductility [18–20].

This work is aimed at investigating the fatigue failure kinetics and mechanism of the VT6 titanium alloy with a UFG structure formed by equal-channel angular pressing (ECAP) in comparison with the initial coarse-grained (CG) state.

EXPERIMENTAL

The object of investigation was VT6 titanium alloy (composition, wt %: 5.95 V, 5.01 Al, and 89.05 Ti) in the initial CG state and in the UFG state after ECAP. The polymorphic transformation temperature in the alloy is tpt = 975 ± 5°C. The initial state of the material is formed by hot rolling of the billets produced by VSMPO-AVISMA, Verkhnyaya Salda. The ECAP technology of the alloy was as follows. A billet 20 mm in diameter and 100 mm in length was subjected to homogenizing annealing. Then, quenching in water from 960°C with holding for 1 h, tempering at 675°C for 4 h, and ECAP at 650°C were performed (route Bс, φ = 120°, number of passages n = 6) [18].

The alloy structure was studied using an optical and scanning electron (JEOL JSM-6390) microscopes (SEM). A fine alloy structure after ECAP is investigated using a JEOL JEM 2100 transmission electron microscope (TEM) at an accelerating voltage of 200 kV. The alloy hardness tests were performed using a Time Group TH 300 hardness meter. Static tension of round samples 5 mm in diameter was performed using a Tinius Olsen H50KT universal testing machine with a velocity of 5 mm/min. Fatigue tests of prismatic samples 10 mm in thickness with a V-like (r = 0.2 mm) stress concentrator were performed at room temperature according to the three-point bending scheme using an Instron 8802 installation at ν = 10 Hz, R = 0.1, and ΔP = 2000–4500 N. The micro-fractographic analysis of cleavages was performed using a Sigma SEM (Carl Zeiss, Germany).

RESULTS AND DISCUSSION

Figure 1a shows the microstructure of the VT6 alloy in the initial state. In this case, the material in the initial billet has a mixed globular-platelet structure typical of hot-rolled rods and included the primary α phase in the form of equiaxial grains 15 ± 5 μm and a region with a platelet (α + β) structure. The volume fraction of a globular component of the α phase is about 65%.

Fig. 1.
figure 1

Microstructure of the VT6 alloy (a) in the CG state (metallography) and (b) in the UFG state (TEM).

Full size image

The alloy thermal treatment and ECAP led to a decrease in size of primary grains of the α phase. The fraction of the UFG structure with an average grain/subgrain size of the α phase of 240 ± 60 nm was about 70% (Fig. 1b). Structural parameters of the VT6 alloy in the initial (CG) state and after ECAP in the UFG state are presented in Table 1.

Table 1.   Structural parameters of the VT6 alloy
Full size table

It is seen from Table 2 that ECAP following the mentioned modes increases the hardness and strength characteristics of the VT6 alloy by a factor of approximately 1.2 but decreases ductility by a factor of 1.4.

Table 2.   Mechanical properties of the VT6 alloy
Full size table

Fatigue tests showed that, at the same load level (ΔP), the fatigue life (Ntot is the number of cycles before failure) of the samples with a stress concentrator in the CG state is higher than after ECAP (Table 3). The number of cycles before the nucleation of the fatigue crack (Nnuc) increases with an increase in sample fatigue life Ntot. Herewith, the value of Nnuc in the samples made of the CG alloy is higher than that of the UFG alloy, although in the percentage, irrespective of the alloy state, Nnuc = 19–23% of Ntot (Table 3).

Table 3.   Total fatigue life of the VT6 alloy samples before the failure and relative number of cycles before the fatigue crack nucleation in them
Full size table
Table 4.   Length of fatigue zone on the cleavage surfaces of the VT6 alloy samples in the initial state and after ECAP
Full size table

Thus, the presence of a stress concentrator in the samples shortens the time before the fatigue-crack nucleation and, consequently, the total material fatigue life, to a larger extent in the UFG state when compared with the initial state.

The straight-linear segment in kinetic diagrams of the fatigue failure of the VT6 alloy depicted in Fig. 2 is approximated by the Paris equation:

$$dl/dN = {C}\Delta {{K}^{n}},$$

where dl/dN is the propagation velocity of the fatigue crack, ∆K is the span of the stress intensity coefficient near the crack vertex, and C and n are coefficients.

Fig. 2.
figure 2

Straight-linear segment of kinetic diagrams of the fatigue failure of the VT6 alloy in the initial state (empty signs) and after ECAP (filled signs).

Full size image

The Paris equation for the CG alloy is depicted as dl/dN = 7.6 × 10–12∆K3.6 and, for the UFG alloy, it has the form dl/dN = 2.5 × 10–11∆K3.8.

It is seen from Fig. 2 that the fatigue-crack propagation velocity in the alloy with an UFG structure is somewhat higher than in the sample with a CG structure at the same ΔK. However, the values of coefficient n in the Paris equation are very close to one another (3.6 and 3.8), which points to the identical sensitivity of alloys to cyclic overloads [20].

Two zones are observed in formed cleavages, notably, the zone of the fatigue-crack development (lf) and rupture zone [20, 21]. It is noteworthy that there is no clearly pronounced boundary between them (Fig. 3). The length of the lf zone, depending on the load, is presented in Fig. 4. It is seen that values of lf are larger for the alloy in the initial state than after ECAP for cleavages formed at the same value of ΔР.

Fig. 3.
figure 3

Appearance of fatigue failures of the VT6 alloy (a) in the initial CG state and (b) after ECAP in the UFG state. (a) ΔP = 2500 and (b) 2000 N. Arrows show the length of the crack fatigue development zone (lf).

Full size image
Fig. 4.
figure 4

Microrelief of the fatigue failure surface of the VT6 alloy in the CG state (magnification 5000×). (a) Microrelief in the failure region (ΔР = 2500 N), (b) and (c) with a fatigue-crack length of (b) 3 mm and (c) 6 mm, and (d) in the rupture zone.

Full size image

Microfractograhic investigations of fatigue cleavages of the VT6 alloy in the CG state showed that the elongation zone is absent in the failure region [21] and cylindrical cleavage facets are observed (Fig. 4a). The microrelief over the entire length of fatigue-crack development can be characterized as scaly (Figs. 4b, 4c). Fatigue grooves are seen on the scale surface. The microrelief of the rupture zone consists of smooth nonuniform pits (Fig. 4d).

A microrelief region 4–6 μm in length can be observed in the failure region of fatigue cleavages of the alloy in the UFG state. Then the microrelief, similarly to the previous case, can be characterized as scaly with fatigue grooves on the scale surface (Figs. 5b, 5c). The rupture zone consists of very fine nonuniform smooth pits (Fig. 5d).

Fig. 5.
figure 5

Microrelief of the fatigue-failure surface of the VT6 alloy in the UFG state (magnification 5000×). (a) Microrelief in the failure region (ΔР = 2000 N), (b) and (c) with a fatigue-crack length of (b) 3 mm and (c) 6 mm, and (d) in the rupture zone.

Full size image

CONCLUSIONS

(i) The fatigue life of VT6 alloy samples with the stress concentrator is higher in the initial (CG) state than after ECAP in the UFG state under the same loading conditions. The number of cycles before the fatigue crack nucleation is at a level of 19–23% regardless of the alloy state.

(ii) The propagation rate of the fatigue crack in the alloy with an UFG structure is somewhat higher than in the CG material at the same ΔK in the Paris equation. However, values of coefficient n for both samples are almost indistinguishable.

(iii) The microrelief of fatigue cleavages of the VT6 alloy both in the CG and UFG state can be characterized as scaly with fatigue grooves on the scale surface. The microrelief of the rupture zone consists of smooth nonuniform pits, regardless of the alloy state.

FUNDING

This study was supported by the Russian Foundation for Basic Research, project no. 15-48-02119 r_povolzh’e_a. RZV gratefully acknowledge the financial support from Saint Petersburg State University in the framework of Call 3 project (id 26130576).

CONFLICT OF INTERESTS

The authors claim that they have no conflict of interests.