Mechanical proper es and deforma on mechanisms in microscale Ti-55531 alloy

The mechanical behaviors and deforma on mechanisms of solu on treated and 450°C aged Ti-55531 alloy microsamples are inves gated by the micro-compression and micro-bending. The results show that the disloca on slip is a concentrated movement in the solu on treated microsamples which containing nanoscale ω phase. Under the tensile and compressive stress, there were serious strain bursts occurred in the plas c deforma on of solu on treated microsamples. But the disloca on slip is transmit at the phase interface in the 450°C aged microsamples which containing high density of α phase. The deforma on behavior exhibits high strength and excellent stability in the micro-compression and micro-bending process. The different mechanical behaviors and deforma on mechanisms of Ti-55531 alloy containing different second phase have significant guidance for tailoring the microstructure and mechanical proper es.

precipita on-strengthening system, since it forms the basis for many types of age-hardening alloys with technological importance [14]. The precipita on strengthening mechanisms can be divided into two categories, the disloca on shearing mechanism and the Orowan disloca on bypassing mechanism, depending on the interac on between moving disloca ons and precipitates. The disloca on-shearing mechanism is generally ac ve when the precipitates are coherent with the matrix, and the par cle size is small, while the Orowan bypassing mechanism dominates when the coherent par cle size exceeds a cri cal value or when the par cles are incoherent with the matrix [4,15,16].
Titanium alloys are considered as a�rac ve materials due to their high strength-to-weight ra o, and abundant phase transi on process under different heat treatment condi ons. Their mechanical proper es are strongly dependent on the microstructural characteris cs, especially the volume frac on, the morphology and the distribu on of the second phase [17][18][19][20], therefore the second phase in β matrix has been a topic of intensive studies. It is worth no cing that the proper es of the β-Ti alloys are almost tested from the polycrystalline bulk materials, and the effect of grain boundaries cannot be ignored. As we known, grain boundaries have significate effect on the mechanical proper es of the metals [21][22][23]. Precise mechanis c analysis using nanoindenta on becomes widely used to inves gate the local area deforma on characters [24,25]. Therefore, the microsamples test was adopted to isola ng individual microstructural units and selec ng par cular slip systems for inves ga on. In this study, the solu on treated and the 450℃ aged Ti-55531 alloy were chosen as the object to inves gate the deforma on mechanisms and the interac on details between disloca ons and different second phases under uniaxial compression and bending.

Methods
Polycrystalline Ti-5wt%Al-5wt%Mo-5wt%V-3wt%Cr-1wt%Zr (Ti-55531) bulk samples with 4 mm ´ 4 mm in cross sec on and 7 mm in length, were solu on heat-treated at 950℃ for 24 h in a high vacuum furnace followed by water quench. One of the bulks needed a further isothermal vacuum annealing, performed at 300 ℃ for 20h to form a large number of athermal ω precipitates [26][27][28]. The annealing temperature was further increased to 450 ℃ for 15h to dissolute the remained ω and promote the growth of α precipitates [27]. The 300°C treatment and the 450 ℃ treatment are associated with the ageing treatment.
The square bulks were mechanically and electrochemically polished before fabrica ng into pillars. For both of the Ti-55531 alloy bulks, grains oriented to <101>β were selected to fabricate micro-can levers. A group of single-crystal micropillars were fabricated using the FEI Helios operated at an ion beam voltage of 30 kV with a variety of currents. Micropillar diameter, ranges from 300 nm to 3 μm, with an aspect ra o of ~2.5. The taper of the micropillars is between 2° and 4°. Three to five samples were prepared for each size. Uniaxial compression experiments were conducted in Hysitron Ti-950 with a 10μm flat punch diamond p at a constant nominal strain rate of 1×10 -3 s -1 . A group of micro-can levers were fabricated using the FEI Helios Dual Beam Focus Ion Beam (FIB) microscope operated at an ion beam voltage of 30 kV with a variety of currents. The dimensions of all can levers are 15 μm long and 2.5 μm wide with an equilateral triangular cross-sec on. Three to five samples were prepared for each size. Micro-bending tests were performed in the Hysitron TI950 nanoindenta on, at a nominal axial strain rate of 1×10 -3 s -1 at the bo�om apex of the can lever at its built-in end [29]. The micro-can lever sample was deflected to 3 μm at the load point, as indicated by the white arrow in the Fig. 1(b), The images of micropillar and micro-can lever are shown in Fig.1. The post-compression morphology of the pillars was examined with scanning electron microscopy (SEM).   Nanoscale ω phase homogeneously were distribute in the β matrix. These ω phase were produced through athermal transforma on from the β to the ω during water quenching.

Results and discussion
For the 450℃ aged Ti-55531 alloy, high density of α phase were formed in the β matrix. The size of α phase was 50-100nm, which was much larger than that of the ω phase. The α phase were short rod-like and distributed uniformly in the β matrix, as shown in Fig. 2 Typical stress-strain curves of the solu on-treated and the 450 ℃ aged Ti-55531 alloy micropillars with various diameters compressed along <101> β orienta on were shown in Fig. 3. The yield strength was defined as the flow stress of the micropillar at 0.2% engineering strain. For the 450 ℃ aged Ti-55531 alloy micropillars, the stress-strain curves were smooth and con nuous, as shown in Fig. 3(a)-(b). The yield strength was about 2.3GPa, regardless of the pillar sizes. For the solu on-treated Ti-55531 alloy micropillars, the strain bursts become more serious for smaller micropillars, as shown in Fig. 3  For the 450℃ aged Ti-55531 alloy micro-can lever, the loading curve was smooth and con nuous, as shown in Fig. 4(a). For the solu on-treated Ti-55531 alloy micro-can lever, the bursts were obvious in the loading curve, as shown in Fig. 4(b). At the same me, the strength of the 450℃ aged Ti-55531 alloy micro-can lever was higher than that of the solu on-treated Ti-55531 alloy micro-can lever. Both the micropillar compression results and the micro-can lever bending results show the excellent stability in plas city of microscale 450℃ aged Ti-55531 alloy.    For the deformed 450℃ aged Ti-55531 alloy micro-can levers, very ny slip traces were observed on the surfaces of pillars rather than the significant slip steps observed in solu on-treated micro-can levers. There were only a few slip traces at the top and the bo�om, as shown in Fig 6(d). All the slip steps were zigzag. Different with the solu on-treated samples, the deforma on in the 450 ℃ aged samples was homogeneous. The slip trace on the surface was mild and shallow. It had found that the interface between the α phase and β matrix was semi-coherent and following the Burgers orienta on rela onship. Compared with the nanoscale ω phase, the size of α phase was much larger, and the disloca on can not directly cut through the α phase.
So the α phase had a significant effect on impeding the disloca on movement. On the other hand, the Burgers orienta on rela onship between the α phase and the β matrix promoted the disloca on slip transmit at the interface. So the disloca on slip system could be ac vated in the α phase. The disloca on "slip relay" model had described in our previous work [34]. The model of disloca on movement was applicable under both compressive and tensile stress. Generally speaking, the nanoscale α phase formed in the ageing treatment is considered as hard par cles in the bulk samples and can not deform in the plas c deforma on, so the disloca on slip can not transfer across the interface easily. But in the microscale samples, the slip systems of 7 MATEC Web of Conferences 321, 11024 (2020) https://doi.org/10.1051/matecconf/202032111024 The 14 th World Conference on Titanium nanoscale α phase can be ac vated by the size effect. The interac on between the disloca on and phase interface significantly improves the yield strength and the plas c deforma on stability.

Conclusion
The results of micro-compression and micro-bending tests show the different mechanical behaviors and deforma on mechanisms for solu on treated and 450℃ aged Ti-55531 alloy samples. The disloca on slip is a concentrated movement in the solu on treated microsamples which containing nanoscale ω phase. Under the tensile and compressive stress, the plas c deforma on of the solu on treated microsamples is unstable because the disloca ons slip in localized area which causes strain bursts. While the disloca on slip is transmi�ed at the phase interface in the 450℃ aged microsamples which containing high density of α phase. The deforma on behavior exhibits high strength and excellent stability in plas city in the micro-compression and micro-bending deforma on.