Sha W., Malinov S. Titanium Alloys: Modelling of Microstructure, Properties and Applications

Подождите немного. Документ загружается.

Synchrotron radiation X-ray diffraction 35

3.2.1 Ti-6Al-4V

Ti 6-4 alloy in five different heat treatment conditions is examined:

(i) annealed (this is the starting state of the alloy, which is rolled with a

reduction degree not less than 60% at temperatures in the α + β field,

followed by recrystallisation annealing at 800 °C for two hours);

(ii) furnace cooling, with a cooling rate of 0.5 °C/s, after β-homogenisation

at 1100 °C;

(iii) water quenching after β-homogenisation at 1100 °C;

(iv) water quenching after homogenisation in α + β region at 850 °C; and

(v) water quenching after homogenisation in α + β region at 850 °C,

followed by ageing at 600 °C for 20 hours.

Ti-6Al-2Sn-4Zr-2Mo-0.08Si Room temperature

30 32 34 36 38 40

2θ (°)

Relative intensity

2.5

1.5

0.5

Water

quenching

Furnace

cooling

Rolled

{100}α′

{002}α′

{101}α′

{101}α

{100}α

{002}α

{110}β

{002}α

{100}α

{110}β

{101}α

3.2

Diffraction patterns at room temperature in the range of 30–40°

for Ti-6Al-2Sn-4Zr-2Mo-0.08Si under different heat treatment

conditions. The intensities are given in relative values. For clarity, the

diffraction patterns are shifted with respect to each other along the

vertical axis.



Titanium alloys: modelling of microstructure36

This most commonly used titanium alloy is classified as an α + β titanium

alloy. The phase composition of the Ti 6-4 alloy after different heat treatments

is mainly α phase (see Fig. 3.1), with a small amount of β phase in the

annealed condition, 5 wt.%.

There is also a small amount of retained β phase after furnace cooling

from the β-region (see the small peak at 35.6° in 2

in Fig. 3.1). During slow

cooling, the diffusional redistribution of the alloying elements leads to

enrichment of the β phase with β stabiliser (vanadium in the case of Ti-6Al-

4V). As a result, a small amount of β phase remains stable at room temperature.

A similar observation for the alloy studied is shown in Chapter 7 for cooling

Beta21s Room temperature

Intensity (counts)

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

30 35 40 45 50 55 60 65 70

2θ (°)

{110}β

{200}β

Water

quenching

{211}β

{110}β

{200}β

{211}β

Furnace

cooling

{100}α

{200}β+{002}α

{101}α

Aged

{102}α

{200}β

{211}β+

{103}α

{112}α

{201}α

3.3

Diffraction patterns at room temperature and profile fits (dotted

lines) in the range of 30–70° 2

for β21s under different heat

treatment conditions. For clarity, the diffraction patterns are shifted

with respect to each other along the vertical axis.



Synchrotron radiation X-ray diffraction 37

from the β-region with different cooling rates. The amount of the β phase

after furnace cooling is about 7 wt.%.

An interesting observation is ascertained for the Ti 6-4 alloy regarding the

reflections at high 2

angles. After furnace cooling, the reflection {103}α

consists of a number of subpeaks (see Fig. 3.4b, ‘Furnace cooling’). This

fact is also observed for {102}α and {112}α reflections, but not for hkl

reflections having l = 0 (see Fig. 3.4a). After slow furnace cooling, α phase

with different c and the same a lattice parameters is formed. The a lattice

parameter of the α phase after furnace cooling is 0.2938 nm, while the c

lattice parameter corresponding to the four subpeaks in Fig. 3.4b is 0.4701,

0.4684, 0.4668 and 0.4655 nm (see Table 3.1). The above observation can be

explained by considering the nature of the β to α transformation in titanium

alloys. For titanium alloys, the transformation from β to α is of monovariant

type. For different temperatures, different amounts of α and β phases are in

Intensity (counts)

1600

1400

1200

1000

800

600

400

200

Ti-6Al-4V Room temperature

Water quenching

Furnace cooling

Annealed

56 56.5 57 57.5 58

2θ (°)

(a)

3.4

Reflections and profile fits (dotted lines) of (a) {110}α and

(b) {103}α for Ti-6Al-4V under different heat treatment conditions. For

clarity, the diffraction patterns are shifted with respect to each other

along the vertical axis.



Titanium alloys: modelling of microstructure38

equilibrium and the α phase can be precipitated at different stages

(temperatures) during slow continuous cooling. The α phase precipitated at

different temperatures has different morphology – finer lamellae are attributed

to lower temperatures of transformation (Chapter 6). This may result in

different levels of the residual stresses for α phase precipitated at different

temperatures. In addition, the composition in terms of alloying elements of

the α phase precipitated at different temperatures is different. The above two

reasons are the most plausible explanation for the α phase observed with

different c lattice parameters.

After water quenching from both β-region (1100 °C) and α + β region

(850 °C), the structure of the Ti 6-4 alloy consists only of hcp α phase. No

amounts of β phase are present. This structure is a product of diffusional

(small fraction in final structure) β to α and diffusionless (large fraction) β

to α′ transformations upon quenching. The diffusionless transformation occurs

at temperatures below 700–750 °C and results in the formation of the martensite

structure (α′). The evidence for the presence of martensite in these cases is

the much wider α reflections (see Fig. 3.5). The full-width half maximum

3.4

Cont’d

62.5 63 63.5 64 64.5 65

2θ (°)

(b)

Ti-6Al-4V Room temperature

Water quenching

Furnace cooling

Annealed

Intensity (counts)

3500

3000

2500

2000

1500

1000

500



Synchrotron radiation X-ray diffraction 39

Table 3.1

Lattice parameters in nanometers at room temperature for titanium alloys after different heat treatments

Alloy Phase Lattice Annealed Furnace cooled Water quenched Water quenched

parameter from β from β from α + β

Ti-6Al-4V α

0.2935 0.2938 0.2935 0.2939

0.4673 0.4701 0.4668 0.4673

0.4684

0.4668

0.4655

0.3226 0.3228

Ti-6Al-2Sn-4Zr-2Mo-0.08Si α

0.2937 0.2942 0.2949

0.4687 0.4698 0.4690

0.3254 0.3264

β21s α

0.2943

0.4686

0.3267 0.3272 0.3266



Titanium alloys: modelling of microstructure40

(FWHM) values of all α reflections are nearly twice those in the furnace

cooled and annealed alloy. It is suggested that, for Ti 6–4 alloy, quenching

from the 750–900 °C temperature range produces an orthorhombic martensite

(α″). This is not confirmed in the diffraction patterns shown here. The diffraction

patterns of Ti 6-4 quenched from 1100 and 850 °C are similar, showing the

presence of hcp martensite only. Ageing at 600 °C of the alloy quenched

from 850 °C leads to small changes in the diffraction pattern (Fig. 3.1, top).

A shoulder appears on the left side of the {101}α reflection. In addition, a

new reflection is observed at 32.75° (d = 0.2483 nm). There is a new phase

precipitated from the α′ phase. It is possible that the orthorhombic martensite

(α″) is precipitated upon ageing, but the reflections are weak.

3.2.2 Ti-6Al-2Sn-4Zr-2Mo-0.08Si

Ti 6-2-4-2 in three different heat treatment conditions is examined, namely

rolled (this is the starting state of the alloy, which was processed from an

ingot at a temperature above the β-transition and then successively α/β

rolled) and furnace cooling and water quenching after β-homogenisation at

1100 °C. The phase composition under all three conditions is mainly α phase

(see Fig. 3.2). Small amounts of β phase are present in the rolled and furnace

cooled alloy, 8 wt.% for both cases. The intensity of the {002}α reflection in

the rolled alloy is significantly higher than the intensities of {100}α and

{101}α reflections due to the presence of crystallographic texture. No retained

FWHM (∆2θ)

0.6

0.5

0.4

0.3

0.2

0.1

{100}α {101}α {100}α {101}α

Ti-6Al-4V Ti-6Al-2Sn-4Zr-2Mo-0.08Si

Reflections

Annealed

Furnace

cooling

WQ(1100)

WQ(850)

Ageing

3.5

Full-width half maximum for α reflections of Ti-6Al-4V and Ti-6Al-

2Sn-4Zr-2Mo-0.08Si under different heat treatment conditions.



Synchrotron radiation X-ray diffraction 41

β phase is present after water quenching. The only phase in the alloy after

water quenching is the hcp martensite (α′) phase (see Figs. 3.2 and 3.5). The

phase constitution of Ti 6-2-4-2 after different heat treatments is similar to

that of Ti 6-4 under the same heat treatment conditions. The lattice parameters

of the α phase in both alloys are similar. The lattice parameters of the β

phase for Ti 6-2-4-2 alloy are appreciably larger than those for Ti 6-4 alloy

(Table 3.1). This effect most probably is due to the different composition of

the β phase at room temperature. In Ti-6Al-4V alloy, the β phase is stabilised

at room temperature as a result of its enrichment with vanadium. More than

15 wt.% vanadium (corresponding to Mo-equivalent [Mo]

of 11 wt.%) is

necessary to stabilise the β phase at room temperature. In Ti-6Al-2Sn-4Zr-

2Mo-0.08Si alloy, the β phase is stabilised at room temperature as a result of

its enrichment with molybdenum. The β phase for this alloy has more than

20 wt.% molybdenum at room temperature (Chapter 6). Since vanadium has

lower atomic radius as compared to titanium it causes a decrease in the

lattice parameter of the β phase in the substitutional solution. Molybdenum,

conversely, with its larger atomic radius, causes an increase in the lattice

parameter.

Considering the schematic pseudo binary phase diagram of titanium alloys,

the Ti 6-2-4-2 alloy is to the left-side of the Ti 6-4 alloy. This implies that the

amount of the residual β phase in Ti 6-4 should be larger than in Ti 6-2-4-2

alloy. The X-ray data indicate that the amount of the residual β phase in Ti

6-4 alloy is slightly lower when compared to the Ti 6-2-4-2 alloy. The reason

for this discrepancy most probably is due to the difference in the oxygen

levels in the two alloys. The amount of oxygen in the Ti 6-4 alloy (0.19

wt.%) is significantly higher than in the Ti 6-2-4-2 alloy (0.065 wt.%).

Oxygen stabilises the α phase. The β phase is therefore depressed in the Ti

6-4 alloy because of the high oxygen content, and promoted in the Ti 6-2-4-

2 alloy because of the low oxygen content.

3.2.3 β21s

β21s in three different heat treatment conditions is examined – aged (this is

the starting state of the alloy, which was processed through β solution treatment

followed by ageing) as well as furnace cooling and water quenching from

900 °C. In the aged alloy, a mixture of α + β crystal structures is present (see

Fig. 3.3). The amount of the α phase is 58 wt.%. The presence of α phase in

the aged condition is a result of the heat treatment, i.e. β solution treatment

followed by ageing. This is a typical condition for a commercial β21s alloy

because α precipitation provides strengthening.

Both water quenching and furnace cooling from 900 °C produce pure β

phase microstructure. α phase is not observed after heat treatment. The

martensite transus temperature in this alloy is lowered by the alloying elements



Titanium alloys: modelling of microstructure42

(15.75% [Mo]

) to temperatures below room temperature. This means that

martensite phase (α′) cannot exist in this alloy. As a result, retention of the

β phase at room temperature is permitted if the β to α phase transformation

is suppressed upon cooling. The α phase does not form even after slow

furnace cooling at a cooling rate around 0.5 °C/s. Since a certain amount of

α phase should be in equilibrium, the reasons for the suspending of the α

phase are kinetic.

The lattice parameters of the β phase in the β21s alloy are larger than

those in Ti 6-4 and Ti 6-2-4-2 alloys (see Table 3.1). Again, the reason is in

the alloy composition (14.1 wt.% Mo, 3 wt.% Al and 3.48 wt.% Nb). Both

molybdenum and niobium have higher atomic radii than titanium, and therefore

they increase the lattice parameter when substituting titanium atoms in the

bcc β phase.

3.3 Measurements at elevated temperatures

Consecutive HR-XRD measurements at different high temperatures with a

number of scans at each temperature are needed in order to reveal the kinetics

of possible phase transitions, for alloys under different heat treatment

conditions. Some of the diffraction patterns obtained from the high temperature

measurements of Ti 6-4, Ti 6-2-4-2 and β21s alloys are shown in Figs. 3.6,

3.7 and 3.8.

The oxygen content of the alloys after high temperature HR-XRD

measurements is increased, to 0.6 wt.%, 0.5 wt.% and 0.8 wt.% for Ti 6-4

(annealed), Ti 6-2-4-2 (rolled) and β21s (furnace cooled), respectively. These

values are the average oxygen contents of the bulk samples. The vacuum

level (0.3 Pa) of the furnace chamber is not sufficient to completely prevent

oxidation.

An increase in the oxygen content raises the β-transus temperature. The

phase equilibria of the alloys at the elevated temperatures are calculated for

different oxygen contents (see Fig. 3.9), with Thermo-Calc software and

using the Ti-database, for oxygen contents in the range from 0 to 1.4 wt.%.

The Ti-database is validated for oxygen concentrations up to 0.3 wt.%.

Therefore, large errors above this value are possible. Indeed, a very significant

influence of the oxygen content on the equilibrium contents of α and β

phases is demonstrated. The increased oxygen content enhances and stabilises

the α phase in respect to the β phase.

The microstructure analysis of cross-sections of Ti 6-4 and Ti 6-2-4-2

after high-temperature HR-XRD measurements using scanning electron

microscopy shows the presence of a surface oxidised layer (see Fig. 3.10).

The microstructure of the surface layer consists of coarse α phase lamellae.

This is a well-defined layer with an obvious boundary between the layer and

the matrix. In addition, some coarse α phase colonies, grown from the surface



Synchrotron radiation X-ray diffraction 43

layer towards the core, are present. The microstructure of the core consists of

very fine α + β plates colonies. The microhardness of the surface layer for

Ti 6-4 alloy is 528 HV1 while the microhardness of the matrix is 382 HV1.

The difference in the microhardness of the surface layer and that of the

matrix is due mainly to the difference of the microstructure. The difference

in the oxygen level may also have an influence on the microhardness difference.

The measurements performed on the cross-sections using scanning electron

microscopy with energy dispersive X-ray and wavelength dispersive X-ray

with an oxygen detection capability of 0.8 wt.% did not detect any change in

the oxygen counts for the surface layer and the matrix. Hence, the difference

in the oxygen levels between the surface layer and the core is small.

Before heating

After heating

Scan 1

Scan 2

Scan 3

Ti 6-4 annealed

× 10

Intensity (counts)

2.5

1.5

0.5

1000 °C

700 °C

600 °C

Room

temperature

{100}α

{002}α

{101}α

{100}α+

{110}α″

{002}α

{020}α″

{101}α

{111}α″

{021}α

Scan 1

Scan 3

Scan 6

{100}α+

{110}α″

{002}α

{020}α″

{101}α

{111}α″

{021}α″

{002}α

{101}α

{100}α

Before heating

After heating

31 32 33 34 35 36 37 38 39

2θ (°)

3.6

Diffraction patterns at different temperatures for Ti-6Al-4V alloy.

For clarity, the diffraction patterns are shifted with respect to each

other along the vertical axis.



Titanium alloys: modelling of microstructure44

The surface layers were formed during the entire cycle of high-temperature

measurements. The surface layer shown in Fig. 3.10 has a thickness of 145–

160 µm, and is a result of high temperature exposure as follows: 1.5 hours

at 600 °C, 1 hour at 700 °C, 0.5 hour at 800 °C, 0.5 hour at 900 °C and 0.5

hour at 1000 °C. Hence, the phase transformations observed at high temperature

are in conditions of concurrent oxidation.

The diffraction patterns are obtained from a surface layer the thickness of

which is limited by the depth of the X-ray beam penetration. The depth of

the X-ray penetration is in the range of 5–6.5 µm for a scan in the range of

30–40° in 2

and 12 µm at 70° in 2

. Hence, the diffraction patterns are

Before heating

After heating

Scan 1

Scan 3

Scan 6

Ti 6-2-4-2 rolled

× 10

Intensity (counts)

4.5

3.5

2.5

1.5

0.5

1000 °C

800 °C

600 °C

Room

temperature

{100}α

{002}α

{101}α

{100}α

{002}α

{101}α

Scan 1

Scan 3

Scan 6

{100}α

{002}α

{101}α

{100}α

Before heating

After heating

31 32 33 34 35 36 37 38 39

2θ (°)

{110}β

{101}α

3.7

Diffraction patterns at different temperatures for Ti-6Al-2Sn-4Zr-

2Mo-0.08Si alloy. For clarity, the diffraction patterns are shifted with

respect to each other along the vertical axis.

