Some data on the influence of structure and chemical composi on on the ra o of fracture toughness and tensile bri�leness of tanium alloys

The development of the aerospace industry is also associated with the development of a complex of mechanical proper es of tanium alloys. The main high-strength pseudo-beta tanium alloys were developed in the 70s-80s,but at the current moment they have not occupied a m ainsegment in applica ons, significantly inferior in terms of the volume to the Ti6Al4V alloy. We can formulate the main requests from the designers, which are rela vely constant: reducing the weight of the structure, increasing strength, increasing s ffness, increasing fracture toughness. If an increase in the strength and fracture toughness of a material is possible and is one of the main tasks of the developers of materials, then an increase in the rigidity of tanium alloys is possible only through the use of composite materials. Reducing the weight of the structure within the framework of tanium alloys (the density is specified) is possible only through a decrease in the cross-sec onal area of the part, which is inversely propor onal to strength, i.e. with increasing strength.


Introduc on
The development of the aerospace industry is also associated with the development of a complex of mechanical proper es of tanium alloys. The main high-strength pseudo-beta tanium alloys were developed in the 70s-80s,but at the current moment they have not occupied a m ainsegment in applica ons, significantly inferior in terms of the volume to the Ti6Al4V alloy. We can formulate the main requests from the designers, which are rela vely constant: reducing the weight of the structure, increasing strength, increasing s ffness, increasing fracture toughness. If an increase in the strength and fracture toughness of a material is possible and is one of the main tasks of the developers of materials, then an increase in the rigidity of tanium alloys is possible only through the use of composite materials. Reducing the weight of the structure within the framework of tanium alloys (the density is specified) is possible only through a decrease in the cross-sec onal area of the part, which is inversely propor onal to strength, i.e. with increasing strength.
In addi on, the increase in strength leads not only to a reduc on in the weight of the final part, but also to a reduc on in the weight of the stock, which ensures a reduc on in cost, which is one of the most actual market demands.
So for parts with a sec on> 1 "when cooled in air rela ve to the base strength level of UTS = 1050 MPa of Ti6Al4V alloy, it is possible to increase the strength on the VST55531 alloy to UTS = 1650 MPa, which is equivalent to a 1.6 mes reduc on of the sec on or 37%. For thin sec ons with the possibility of quenching in water, the equivalent reduc on in the sec on can be 1.4 mes or 28%.
Thus, an increase in the strength of tanium alloys is a very promising and realiable direc on, which, however, has several limita ons.
With an increase in strength for all classes of tanium alloys, duc lity under tension and fracture toughness decrease. Upon reaching the cri cal strength at which the failure occurs immediately a�er the elas c deforma on, without any plas c deforma on. That is plas city (rela ve elonga on and reduc on area) under tension = 0. The material in this state is a tensile macro bri�le. Similarly, at a certain strength, the fracture toughness may decrease to 0. This state will be called «toughness bri�le».
The plas city realized by the material during plas c deforma on depends on the stress-strain state. In order of increasing severe of loading: compression, torsion and tension. Tension can be considered the most severe loading at which the minimum plas city resource is realized. Therefore, in tensile bri�le states, true strength is as close as possible to the test for compression or hardness.
The danger of bri�leness lies in the unpredictability of the results of its loading. When the threshold of bri�le tensile strength is exceeded, with a further increase in the true strength (determined by hardness), an increase in tensile strength is observed in a small area (Fig. 1). With further increase in true strength, maximum tensile strength decreases. And there is a range of UTS, in which failure can occur with a certain probability. The greater the difference between the true strength and the threshold bri�leness strength, the greater the range of possible UTS (Fig. 1).
The dimly lit ques on remains the ra o of tensile bri�le and toughness bri�le strengths, as well as the impact on this ra o of chemical composi on, structure parameters, crystallographic texture etc. There is li�le systema c experimental data because the main task is to avoid bri�leness states.

Material and experiments
The work carried out rods deformed in a + b-field, chemical composi on is given in table.1.
The tensile test was carried out on samples of Ø6.0 mm with the gage length 4D according to ASTM E8. The tensile was carried out on an MTS Z100 tes ng machine with a nominal force of 10 kN at room temperature. Impact toughness was tested on samples with a cross sec on of 10x11mm with a V-shaped notch and a pre-grown fa gue crack. The torsion test was carried out on samples of Ø10mm and a working part length of 50mm, on a universal tes ng machine TNS-DW 05 with a nominal torque of 500 N*m. The microstructure was inves gated on microslises a�er prepara on on a composite wheel and final polishing using Colloidal Silica of 0.05 μm dispersion. The microstructure and surface damage was inves gated on a FEI Quanta 3D FEG and SEM Zeiss Sigma 300 VP scanning electron microscope with an accelera ng voltage of up to 30 kV.

Results and discussion
The work was carried out on alloys, with significant hardening due to precipita on age hardening.
For the experimental alloy of the martensi c class (Table 1), the strength of the aging regime was varied in different ini al states. At hi-temperature aging (HTA), minimal hardening was realized due to large precipita ons of the secondary phases. At low temperature aging (LTA), the maximum hardening was realized due to the fine discharge of the secondary phases.
During beta quench α "martensi c transforma on occurs. Low temperature aging (LTA) results in a very fine mixture of α and β phases (Fig. 3 d), which supply increase in material strength up to 460 HV leads to tensile bri�leness (Fig. 2 a), however, the impact strength does not fall to zero. The coarsening of α and β phases a�er hi-temperature aging (HTA) (Fig. 3 a) reduces hardness to 354 HV, and the material is not tensile bri�le, i.e. shows significant elonga on and RA, also showing a twofold increase in impact toughness (Fig. 2b). The magnitude of microplas c deforma on, determined by the local deforma on of the edges of the dimples and fracture facets, decreases during the transi on to the tensile bri�le condi on, but does not undergo qualita ve changes -remains micro duc le (Fig.3 a, d). An atypical feature is the grain boundary fracture in the HTA condi on, which is less energy-intensive compared to the predominantly intragranular fracture in the tensile bri�le LTA condi on.
The transi on to a globular structure with about 20% of the primary globular α-phase a�er α + β quench to metastable allows for rela vely high duc lity (EL up to 12.5 and 19.2%) with comparable hardness, but significantly lower impact toughness in the corresponding HTA and LTA condi ons compared to beta quench condi on. Tensile bri�le beta quench LTA condi on has a significantly higher impact toughness compared to α + β quench LTA condi on, which indicates that there is no unambiguous rela onship between the tensile bri�le and level of impact toughness.
The lamellar structure a�er beta anneal with a close frac on of the primary α-phase and quenching to the metastable bphase reveals the same tendencies as the beta quench condi on (with the transforma on of α "martensite) (Fig. 2). The magnitude of microplas c deforma ons in the tensile bri�le condi on a�er LTA is not significantly different from microplas c deforma on in HTA condi on with significant macro-plas c deforma on under tension (Fig. 3c, f). The level of impact toughness also does not fall to 0 in the tensile bri�le condi on a�er LTA (Fig. 2 b) and corresponds to the corresponding states in the beta quench condi on. In general, the classic picture is observed on this experimental alloy -the lamellar structure has a significantly lower duc lity and a higher fracture toughness compared to the α + β structure.  For a somewhat less alloying VST2 alloy (Table 1), the quenching temperature varied the propor on of the primary αphase (γα1) on the α + β processed structure and beta anneal structure (Fig.4.5), followed by thermo strengthening by aging in one mode. All states with a lamellar structure a�er beta anneal were tensile bri�le. All α + β processed condi on had non-zero elonga on and RA (Fig.4 b). At the quenching temperature of the corresponding γ α1 , up to 18% inclusive, the alloy is quenched by α "martensite. With a larger γ α1 -on the beta phase.
States with the same γ α1 had very close strength, based on hardness data. At the same me, as in the experimental alloy, the difference in the magnitude of the microplas c deforma on of the fracture relief is not significantly different between the beta anneal (tensile bri�le) and α + β processed condi on (Fig. 5 a, d, e, f). At maximum γα1 = 40%, the difference in macroplas c deforma on between the α + β processed and beta anneal condi ons is maximal and significantly larger than the state with γ α1 = 6% (Fig.4 b), however the difference between the magnitude of microplas c deforma on does not increase and remains close.
When γα1 = 6%, the impact toughness has close values for both ini al states (Fig. 4). When γα1 = 18%, the impact toughness in the tensile bri�le beta anneal condi on is lower than in the α + β processed condi on, which is atypical. When γα1 = 40% in the beta anneal condi on has a greater impact toughness, despite the tensile bri�leness, compared with the α + β processed condi on., It is difficult to talk about the reasons for the change in the ra o of impact toughness with γα1 = 18% rela ve to γ α1 = 40%. Most likely the cause is a change in the frac on of the primary a-phase, since the phase composi on a�er quenching for γ α1 = 6 and 18% does not change, but this also causes a change in the ra o of impact toughness.
This alloy also shows the absence of a direct link between the level of impact toughness values and the transi on to tensile bri�leness. The level of impact toughness depends in a complex way on the type of structure and the propor on of the primary α-phase.
In the pseudo-beta VT22 alloy, the strength varia on was made due to the change in the aging temperature (Fig. 7). LTA and HTA were selected so that the LTA state was tensile bri�le, and the HTA was duc le. The difference in the strengths of the LTA and HTA states was maintained at a minimum and was 18-68 HV (80-250 MPa in terms of UTS) (Fig. 6). The duc lity of the material was recorded by tensile and torsion tests. The primary α-phase volume frac ons for quenched condi ons from the α + β region was about 12%.

Fig.6
Mechanical proper es of the VT22 alloy rod a�er aging at high and low temperatures with different types of ini al structure: α + β quench; beta anneal + α + β quench; beta quench.

Fig. 7
Microstructure of the VT22 alloy rod a�er aging a, b, c) at high temperature; d, e, f) aging at low temperature, with a different type of ini al structure:a,d) α+β quench; b,e) beta anneal +α+β quench; c,f) beta quench.
The transi on to a lamellar structure using beta anneal leads to an increase in β-grain up to 180 μm (Fig.7) and a decrease in the strength of the transi on to tensile bri�leness (Fig.6). Impact toughness for LTA is slightly lower than HTA and more than 2.7 mes the α + β quench condi on (Fig. 6). During torsion, the material experiences plas c flow, including in the tensile bri�le LTA condi on, although the deforma on of the torsion LTA is less than the HTA state.
Quenching from the beta region eliminates the primary α-phase and leads to an increase in the β-grain (Fig. 7). Impact toughness for LTA is slightly lower than HTA and almost 2 mes more than beta anneal condi on ( fig.6). As in the beta anneal condi on during torsion, the material get plas c flow, including in the tensile bri�le LTA condi on, although the torsion deforma on of LTA is less than the HTA state.
In general, for the VT22 alloy there is a classic dependence of greater impact toughness and less plas city for the lamellar structure compared to the globular structure.
The coincidence of the bri�le state under tension and torsion for the α + β quench condi on, as well as the presence of plas city under torsion in other bri�le states, confirms the most stringent condi ons from the point of view of the implementa on of plas c deforma on are realized under tension.
Analyzing the general impact toughness behavior tendencies near the tensile bri�le duc le transi on on different classes of alloys and microstructures, we can say that the tensile bri�le duc le transi on is not related to the toughness bri�le transi on, and that the impact toughness func on does not have discon nui es during the bri�le duc le transi on.
The dependence and behavior of impact toughness at tensile bri�leness is complex and depends on a large number of factors, but one of the most significant influense is the type of the primary α-phase structure and its specific frac on. It should be borne in mind that the behavior of impact toughness is very dependent on chemical composi on of the alloy and is unlikely to be versifica on.