Version of Record: https://www.sciencedirect.

com/science/article/pii/S0921509320310790
Manuscript_db3544122e47a1c402ce1889dcbe5704

Fracture toughness of Ti-22Al-25Nb alloy at room and high

temperatures
Penghui Zhanga,b,c*, Weidong Zenga,b,c, Youping Zhenga,b,c,

Jianwei Xua,b,c, Xiaobo Liangd, Yongqing Zhaoa,e

a. State Key Laboratory of Solidification Processing, School of Materials

Science and Engineering, Northwestern Polytechnical University, Xi’an 710072,
China
b. Defense Technoloies Innovation Center of precision forging and ring rolling,
School of Materials Science and Engineering, Northwestern Polytechnical University,
Xi’an 710072, China
c. Shaanxi Key Laboratory of High-Performance Precision Forming Technology
and Equipment, Northwestern Polytechnical University, Xi'an, 710072, PR China
d. Central Iron and Steel Research Institute, Beijing, 100081, PR China
e. Northwest Institute for Nonferrous Metal Research, Xi'an 710016, China
*Corresponding author
E-mail: [email protected]
Abstract: In this research, the fracture toughness of Ti-22Al-25Nb orthorhombic
alloy with bimodal size lamellar O phase microstructures was investigated at room
temperature and 650℃. In addition, the fracture toughness model and crack tip plastic
zone model were established by simplifying the shape of the crack tip plastic zone.
The fracture toughness model is used to quantitatively calculate the intrinsic and
extrinsic contributions to fracture toughness, and the crack tip plastic zone model is
used to calculate the real height of the crack tip plastic zone. The results show that the
fracture toughness KIC at high temperature is twice that at room temperature. The
fracture mechanisms at room and high temperatures are the quasi-cleavage fracture
and dimple fracture, respectively. The increase of temperature is conducive to the
significant increase of the fracture toughness by improving the inherent property of
the material (intrinsic contribution) and the crack propagation tortuosity (extrinsic
contribution). The fracture toughness at room temperature is dominated by intrinsic

© 2020 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
contribution while the fracture toughness at high temperature is dominated by both the
intrinsic and extrinsic contributions. Besides, whether at room temperature or high
temperature, the intrinsic contribution is the most important factor to the total fracture
toughness. The increase of temperature significantly improves the size of the crack tip
plastic zone, which leads to a significant increase in fracture toughness.

Key words: Ti-22Al-25Nb alloy; fracture toughness; fracture mechanism; fracture

toughness model; crack tip plastic zone model

Nomenclature

R the crack propagation tortuosity E the Young's modulus

the real length of the crack

L ν the Poisson's ratio
propagation path
the straight length of the crack intrinsic contribution to fracture
L0
propagation path toughness
extrinsic contribution to fracture
KIC fracture toughness
toughness
the strain at uniform
GIC the critical energy release rate εu
deformation stage
the surface energy per unit
γs yield strength
apparent macro metric area
the energy consumed in plastic
γp O orthorhombic
deformation of crack surfaces
Central Iron and Steel Research
γeff the effective surface energy CISRI
Institute
the specific energy of
Av CT compact tension
deformation
the volume responsible for
V SEM scanning electron microscopy
failure

F the surface of the crack forming IPP6.0 Image Pro Plus 6.0

the real height of the crack tip

l BSE back-scattered electron
plastic zone
the minor axis of the simplified
h CTPZ crack tip plastic zone
ellipse
1. Introduction
The ordered orthorhombic phase (O phase) Ti2AlNb was firstly found in
Ti3Al-Nb alloy in 1988 [1]. Since then, many scholars around the world have carried
out a lot of research. Ti2AlNb based alloys have been widely applied in aerospace
filed due to the merits of high strength, toughness and creep resistance [2-6].
Meanwhile, the damage tolerance design concept has been widely accepted in
aerospace filed [7, 8]. Fracture toughness, an important indicator of the damage
tolerance design concept, will directly affect the life of the aircraft [9]. However, as an
intermetallic material, Ti2AlNb based alloys usually have intrinsic brittleness, which
leads to a significant reduction in fracture toughness and limits its application [10].
Therefore, it is very necessary to study the fracture toughness of Ti2AlNb based
alloys.
Over the years, great efforts have been conducted to research fracture toughness of
intermetallic materials. Eck and Ardell reported that the fracture toughness of
Ti-46.5Al-2.1Cr-3.0Nb-0.2W alloy with duplex microstructure was measured to be
11.93±2.18 MPa·m1/2 [11]. The fracture toughness of Ti-45Al-2Nb-1.5V-1Mo-0.3Y
alloy was investigated by Chen et al, the results showed that the alloy with fine
lamellar microstructures possesses a higher KIC value of up to 23.5 MPa·m1/2 at room
temperature [12]. The high temperature fracture toughness of TiAl alloys with high
Nb content was studied by Yu et al [13], and the results showed the fracture toughness
of the nearly and fully lamellar TiAl alloy with high Nb content at 750℃ were 19.54
MPa·m1/2 and 31.58 MPa·m1/2, respectively. Obviously, the fracture toughness of
these alloys is relatively low and difficult to meet the application requirements [14].
According to the research of Zheng et al, the fracture toughness of Ti-22Al-25Nb
alloy with bimodal size lamellar O phase microstructures can be as high as 30.1
MPa·m1/2 at room temperature, which has great potential in the aerospace filed [15]. It
is well known that many structural components in aerospace application are usually
used in high temperature environments [16]. But it is a pity that the fracture toughness
of Ti2AlNb based alloys at high temperature is rarely reported. Therefore, it is very
urgent to study the fracture toughness of Ti2AlNb based alloys at room and high
temperatures.
The fracture toughness of Ti-22Al-25Nb alloy with bimodal size lamellar O phase
microstructures is studied at room temperature and 650℃ in this paper. The fracture
surface and the crack propagation path are also analyzed. In addition, the fracture
toughness model and crack tip plastic zone model are established by simplifying the
shape of crack tip plastic zone.
2. Experiment and material
The Ti-22Al-25Nb alloy in present paper was provided by Central Iron and Steel
Research Institute (CISRI). The chemical composition of the as-supplied
Ti-22Al-25Nb alloy was Ti-22.3Al-25.7Nb-0.0004O-0.00005N (at.%). The B2
transus temperature of this alloy was 1060℃ [15]. The as-supplied Ti-22Al-25Nb
alloy bar was isothermally forged along the axial direction above B2 transus
temperature and air cooled to room temperature. Then the billet was treated at 960℃
for 1h and water quenched. And the aging treatment was carried out at 780℃ for 12h
and air cooled. Finally, the standard compact tension (CT) sample used in present
paper was processed. Fig. 1 illustrates the geometric dimension of the CT sample.
Pre-cracks were firstly prepared at room temperature by a fatigue machine. The
loading direction and crack propagation direction were shown by the black arrows.
Then the CT samples were tested at room temperature and 650℃ to obtain the
fracture toughness KIC, respectively. Microstructure and fracture surface were
performed by TESCAN VEGA 3 LMU scanning electron microscopy (SEM) [17].
The crack propagation paths were analyzed by optical microscope and the
professional software Image Pro Plus 6.0.
 Fig. 1. The geometric dimension of the CT sample
3. Results and discussion
3.1 Microstructure and fracture toughness KIC
Fig. 2 shows the microstructure of Ti-22Al-25Nb alloy in the back-scattered
electron (BSE) image with the magnification of 5000×. Generally, the light, gray and
dark regions in the BSE images for Ti2AlNb based alloys correspond to the B2, O and
α2 phases, respectively [4,18]. Therefore, it can be found from Fig. 2 that there are
only B2 and O phases in the microstructure of Ti-22Al-25Nb alloy, and α2 phase is not
observed. The present microstructure consists of B2 matrix, lamellar O phases and
acicular O phases. According to the previous research, lamellar O phases are formed
in the solution treatment process and acicular O phases are formed during the aging
treatment process, so this microstructure is called bimodal size lamellar O phase
microstructures [15,19-20]. Furthermore, the fracture toughness KIC of such
microstructure at room temperature and 650℃ was obtained to be 32.5 MPa·m1/2 and
65.4 MPa·m1/2 by experiment, respectively. It can be found that the fracture toughness
at 650℃ is almost twice that at room temperature.
Why is there such a significant difference in fracture toughness between room
temperature and high temperature? In order to find out the reason, the influence
factors of the fracture toughness and their contribution to fracture toughness have to
be considered. Ritchie and Launey believed that the fracture toughness can be
considered to be the result of intrinsic and extrinsic mechanisms, in which intrinsic
mechanism was an inherent property of the material and extrinsic mechanism
depended on crack size (and to some degree specimen geometry) [21,22]. Lütjering et
al thought that the fracture toughness was the result of the intrinsic fracture toughness,
which was the resistance to fracture (ductility) and the extrinsic fracture toughness,
which was the crack front geometry [23]. Shi et al. [24] and Eylon et al. [25]
sustained that the extrinsic fracture toughness can be directly reflected by the crack
propagation tortuosity, and thought that high ductility of the material as well as
tortuous crack propagation path resulted in the highest fracture toughness. Based on
the above theories, the fracture toughness of the material can be considered to be from
two major contributions: intrinsic contribution and extrinsic contribution. The
intrinsic contribution is related to the inherent property of the material; while the
extrinsic contribution is related to the crack propagation tortuosity. The higher the
inherent property of the material and the crack propagation tortuosity are, the higher
the fracture toughness is. Consequently, the intrinsic and extrinsic contributions to
fracture toughness at room temperature and 650℃ are analyzed in detail below.

Fig. 2 The microstructure of Ti-22Al-25Nb alloy in the BSE image

3.2 Fracture mechanism at room temperature

Fig.3 (a) shows the macro fracture surface of the CT sample at room temperature.
The whole fracture surface is composed of the pre-crack zone and crack propagation
zone, and the boundary of the two zones is clear, which is shown by a yellow dotted
line. Through careful observation of the crack propagation zone, the plastic
deformation trace and the shear lip are not observed on the macro fracture surface.
However, a large number of flashy facets are distributed on the fracture surface. These
flashy facets are a typical feature of the cleavage fracture, which belongs to the
transgranular fracture [26-28].
Figs. 3(b)-(e) show the micro features of the crack propagation zone at different
magnifications. As shown in Fig. 3(b), a lot of cleavage facets can be clearly
displayed and look flat at the magnification of 30×. In fact, these cleavage facets are
flashy facets observed on the macro fracture surface, which is the feature of B2 grain
cleavage fracture. And because the Ti-22Al-25Nb alloy bar was compressed during
isothermal forging, these cleavage facets of B2 grains are elongated. In addition,
cleavage steps and cleavage river patterns are exhibited on the fracture surface. Fig
3(c) clearly presents the feature of cleavage steps on the cleavage facets at the
magnification of 200×. It can be seen that cleavage facets become not flat compared
with that of low magnification. Fig. 3(d) shows the cleavage facets deflected by a
certain angle in the process of crack propagation. The more subtle features on the
cleavage facets are shown in Fig 3(e) with the magnification of 5000×. Many
microvoids and bright tearing ridges on the cleavage facets can be clearly observed.
Moreover, some tiny facets with strip shapes are surrounded by tearing ridges and
many microvoids are distributed on tearing ridges. These subtle features present the
feature of ductile fracture in the process of crack propagation. According to Zheng et
al, the features of microvoids and tearing ridges at room temperature are also
observed on the fracture surface of the CT sample of Ti-22Al-25Nb alloy [15]. From
the above, the features of B2 grain cleavage facets and tearing ridges on the fracture
surface indicates the fracture mechanism of Ti-22Al-25Nb alloy with bimodal size
lamellar O phase microstructures is a qusi-cleavage fracture at room temperature.
For Ti2AlNb based alloy, B2 phase is a ductile phase and O phase is a brittle phase.
O phase usually occurs cleavage fracture at room temperature, thus forming tiny
facets observed on fracture surface [29,30]. While B2 phase can blunt the crack and
delay crack growth, thus forming bright tearing ridges and presenting ductile features
[15,29]. It can be found that the presence of B2 phase increases the ability of the
material to resist crack propagation and maintains the inherent property of the
material through the above analysis of fracture features. In other words, the intrinsic
contribution to fracture toughness at room temperature is mainly provided by B2
phase.
In addition, the extrinsic contribution to fracture toughness is studied by the crack
propagation path and crack propagation tortuosity. In order to obtain the crack
propagation path, a small sample along the crack propagation direction was cut from
the fracture surface. The optical microscope image of the crack propagation path can
be explicitly seen in Fig. 4(a). It can be found that the crack is deflected along the
crack propagation direction, which is marked by the red dotted circle in Fig. 4(a). The
feature of the crack deflection indicates that the crack is hindered in the process of the
crack propagation. However, in terms of the whole crack propagation path, the
number of the crack deflection is relatively small and the crack propagation path is
also relatively straight, which indicates that the crack propagation resistance is small.
In other words, the crack can be resisted and the resistance is small in the whole
process of the crack propagation.
Fig. 3. The fracture surface of the CT sample at room temperature. (a) macro fracture
surface. (b), (c), (d) and (e) crack propagation zone. The magnification of (b), (c), (d)
and (e) is 30×, 200×, 1000×, and 5000×, respectively.

Meanwhile, the crack propagation tortuosity is analyzed based on the crack

propagation path. In general, the crack propagation tortuosity can be expressed by
Equation (1) [24,31].

(1)
Where R is the crack propagation tortuosity, L and L0 are the real and straight
lengths of the crack propagation path (as shown in Fig. 4(b)), respectively. It is found
that the real length of the crack propagation path L is irregular.
The crack propagation tortuosity R can be obtained by a professional graphic
processing software named image pro plus 6.0 (IPP6.0). Firstly, the crack propagation
path in Fig. 4(a) was extracted by IPP6.0 software, the result was shown in Fig. 4(b).
According to Fig. 4(b), the real length L and the straight length L0 can be measured by
the automatic scanning function of IPP6.0 software. Finally, the crack propagation
tortuosity R can be obtained.
The R value calculated by the above method is 1.08 at room temperature. It is
found that the R value is very low, which indicates that the extrinsic contribution to
fracture toughness is very small. In fact, the fracture mode of B2 grains is cleavage
fracture at room temperature. Therefore, in the process of crack propagation, the crack
always passes through the B2 grains in a straight manner, resulting in a low crack
tortuosity.

Fig. 4. The crack propagation path at room temperature (a) the optical microscope
image (b) the image processed with IPP 6.0 software

3.3 Fracture mechanism at high temperature

The macro fracture surface of the CT sample at 650℃ is shown in Fig.5(a). It can
be observed that the surface of the pre-crack zone is shallow bule while the crack
propagation zone is dark brown, which is caused by different oxidation time at 650℃
[32]. The crack propagation zone on fracture surface at 650℃ is undulating, which is
different from B2 grain cleavage facets of the fracture surface at room temperature. In
addition, the shear lip along the crack propagation direction can be clearly observed,
which indicates that the fracture mechanism at 650℃ is ductile fracture. Figs.5 (b1)
and (b2) show the SEM images of the crack propagation zone. A large number of
equiaxed dimples can be observed on the crack propagation zone at the magnification
of 1000×, which is quite different from the B2 grain cleavage facets observed at room
temperature. Besides, many microvoids can also be seen in the SEM images with the
magnification of 5000×. And it can be found that the feature of tiny facets at room
temperature is not observed, indicating that the fracture mode of O phase at 650℃ is
not cleavage fracture. Meanwhile, the SEM images of the shear lip zone on the
fracture surface are shown in Figs.5 (c1) and (c2). Plenty of shear shallow dimples are
distributed on the shear lip zone. These shear shallow dimples are stretched along the
direction of shear fracture, which is a typical feature of ductile fracture. From the
above, it is found that the fracture mechanism of Ti-22Al-25Nb alloy with bimodal
size lamellar O phase microstructures at 650℃ is dimple fracture, which belongs to a
typical ductile fracture.
Obviously, the fracture mechanism of Ti-22Al-25Nb alloy with bimodal size
lamellar O phase microstructures at room temperature and 650℃ has a significant
difference. In other words, the increase of temperature changes the fracture
mechanism of Ti-22Al-25Nb alloy with bimodal size lamellar O phase
microstructures, that is, from quasi-cleavage fracture at room temperature to dimple
fracture at 650℃. According to the previous research, at room temperature, only {110}
< 111 > and {112} < 111 > slip systems of B2 phase are activated, and the number of
slip systems of O phase is lack and only ‘a’ dislocation slip can be activated [33,34].
However, {123} < 111 > slip systems of B2 phase and ‘c+a’ dislocation slip of O
phase can be also activated at high temperature [34,35]. Therefore, the increase of
active slip systems at high temperature enhances the inherent property of the material
and improves the resistance of crack propagation. In other words, the intrinsic
contribution to fracture toughness at high temperature is significantly improved.
Fig. 5. The fracture surface of the CT sample at 650℃. (a) macro fracture surface. (b1)
and (b2) crack propagation zone. (c1) and (c2) shear lip zone. The magnification of
(b1) and (c1) is 1000×. The magnification of (b2) and (c2) is 5000×.

In addition, the crack propagation path and crack propagation tortuosity at 650℃
are also analyzed. Fig.6(a) shows the image of the crack propagation path at 650℃. It
can be observed from Fig.6 (a) that the crack propagation path greatly fluctuates along
the crack propagation direction and the number of the crack deflection is large (the
red dotted circle in Fig. 6(a)), which indicate that the crack propagation resistance is
large in the whole process of the crack propagation. Therefore, the crack propagation
path at 650℃ is more tortuous than that at room temperature. Besides, the crack
propagation path at 650℃ was extracted by IPP 6.0 software, as shown in Fig. 6(b).
Then the crack propagation tortuosity R calculated by Equation (1) is 1.45 at 650℃. It
is found that the R value is high, which indicates that the extrinsic contribution to
fracture toughness is large. Meanwhile, the R value at 650℃ is higher than that at
room temperature. That is to say, the extrinsic contribution to fracture toughness at
650℃ is also significantly improved.

Fig. 6. the crack propagation path at 650℃. (a) the optical microscope image (b) the
image processed with IPP 6.0 software

3.4 Establishment and analysis of the model

The fracture mechanism of Ti-22Al-25Nb alloy with bimodal size lamellar O
phase microstructures at room and high temperatures was discussed in the above
sections 3.2 and 3.3, respectively. It can be found that the increase of temperature not
only changes the fracture mechanism from quasi-cleavage fracture at room
temperature to dimple fracture at high temperature, but also changes the crack
propagation path from straight at room temperature to tortuous at high temperature,
indicating that the increase of temperature improves the inherent property of the
material and the crack propagation tortuosity. Consequently, the fracture toughness at
650℃ is much higher than that at room temperature. In other words, the increase of
both the intrinsic and extrinsic contributions to fracture toughness at 650℃ results in
the higher fracture toughness. However, the magnitude of the intrinsic and extrinsic
contributions to fracture toughness is not clear, and which contribution dominates the
fracture toughness of the materials is also not clear. To explain the phenomenon, a
fracture toughness model considering the intrinsic and extrinsic contributions is
established in this section. And then based on the fracture toughness model, a crack
tip plastic zone model is derived.
3.4.1 Fracture toughness model
Fracture toughness is the ability of a material to resist crack propagation. It can be
determined by the critical energy release rate GIC of unstable crack propagation under
plane strain conditions. According to the theory of linear elastic fracture mechanics,
the relation between the plane strain fracture toughness KIC and the critical energy
release rate GIC can be described as follow [36]:
GIC
= (2)
(1-ν2 )

Where E is Young's modulus and ν is the Poisson's ratio. According to

Griffith-Orowan-Irwin theory [37], the critical energy release rate GIC is given as the
following Equation.
GIC=2(γs+γp)=2γeff (3)
Where γs is the surface energy per unit apparent macro metric area which is
necessary to produce unstable crack propagation at the crack tip, γp is the energy
consumed in plastic deformation of crack surfaces and γeff is the effective surface
energy. Equation (3) is usually applied to plastic metallic materials where γp is
significantly considerable.
According to Ragozin et al, the relationship between the specific energy of
deformation Av and the effective surface energy γeff have been established by the
Equation (4).
Av·V=2γeff·F (4)
Where V is the volume responsible for failure, and F is the surface of the crack
forming [36].
 It should be noted that both the volume responsible for failure V and the surface of
the crack forming F are related to the shape of the crack tip plastic zone. Therefore,
the shapes of mode I crack tip plastic zone calculated by theoretical model (top) and
finite element model (bottom) are given in Fig. 7(a) [38]. It can be found that two
shapes is similar to the peanut (top) and broad bean (bottom), which are both irregular
shapes. These irregular shapes will bring great difficulties to the calculation of V and
F.
For the convenience of calculation, we propose an approach to simplify the shape
of mode I crack tip plastic zone in this paper. The above two shapes can be simplified
to be the ellipse shape based on two principles: one is that the original area of the
crack tip plastic zone is equal to the area of the simplified ellipse shape, and the other
is that the height of the crack tip plastic zone l is equal to the major axis of the ellipse
l. In the fracture toughness test, the core of the CT sample is in plane strain state and
the surface is in plane stress state. Therefore, according the simplified method, the
three-dimensional shape of the crack tip plastic zone in the entire thickness direction
of the sample is similar to a waist drum shape. Based on the above simplified method,
for the differential plastic zone with a thickness of db at the core of the sample, the
shape of the crack tip plastic zone can be regarded as an elliptic cylinder, as shown in
Fig. 7(b). Then the volume responsible for failure V and the surface of the crack
forming F can be given by the Equation (5).
πlh
V= db (5)
4
F=h·db
Where l and h correspond to the major axis and minor axis of the ellipse,
respectively. It is worth noting that based on two principles of simplification, there is
no error between the volume of the simplified crack tip plastic zone calculated in
Equation (5) and the volume of the original crack tip plastic zone, and the major axis
of the simplified ellipse l is the real height of the original crack tip plastic zone l .
 Fig. 7. The shapes of crack tip plastic zone (CTPZ) [38]: (a) CTPZ from theoretical
and finite element calculation and (b) simplified elliptic cylinder CTPZ

Then by introducing Equation (5) into Equation (4), Equation (4) can be changed
as follows

2 = 4 Av
πl
(6)

It can be found that the minor axis of the ellipse h is not included in Equation (6),
only the height of the crack tip plastic zone l is included, which indicates the minor
axis of the ellipse h is only a theoretical parameter and cannot be calculated.
Furthermore, combining Equations (2), (3) and (6), Equation (2) can be changed into
πlEAv
= 4(1-ν2 )
(7)

It can be found that there are four main parameters in Equation (7): the height of
the crack tip plastic zone l, the Young's modulus E, the Poisson's ratio ν and the
specific energy of deformation Av. These parameters are all related to the inherent
property of the material. In other words, Equation (7) only considers the intrinsic
contribution to fracture toughness ( ). In fact, the energy consumed in the process
of the crack propagation is not only related to the specific energy of deformation Av,
but also related to the crack propagation path. It should be noted that the crack
tortuosity R has been established by the crack propagation path in Section 3.2.
Therefore, the crack propagation tortuosity R is introduced into Equation (7) [24], a
fracture toughness model can be established as follows.
πlEAv
= R (8)
4(1-ν2 )

Equation (8) has considered the intrinsic contribution (the inherent property of the
material) and the extrinsic contribution (the crack propagation tortuosity) to fracture
toughness. In order to clearly distinguish the intrinsic ( ) and extrinsic ( )
contributions to fracture toughness, Equation (8) can be changed into

+ (√ − 1)
πlEAv πlEAv
= + = (9)
4(1-ν2 ) 4(1-ν2 )

Based on a series of deductions above, a fracture toughness model is established

to quantitatively calculate the intrinsic and extrinsic contributions to fracture
toughness. Table 1 shows the values of the intrinsic contribution and the
extrinsic contribution to fracture toughness by the calculation of Equations (8)
and (9). The values of obtained by experiments are 32.5MPa·m1/2 and
65.4MPa·m1/2 at room temperature and 650℃, respectively. The values of R obtained
by Equation (1) are 1.08 and 1.45 at room temperature and 650℃, respectively. Then
πlEAv
the values of 4(1-ν2 )
at room temperature and 650℃ can be calculated by Equation

(8). Finally, the and at room temperature and 650℃ can be calculated by
Equation (9).

Table 1 The values of and at room temperature and 650℃

T Percentage Percentage
R
(℃) (MPa·m1/2) (MPa·m1/2) of (MPa·m1/2) of
20 32.5 1.08 31.7 97.5% 0.8 2.5%
650 65.4 1.45 54.3 83.0% 11.1 17%

As shown in Table 1, the at room temperature and 650℃ are 31.7MPa·m1/2

and 54.3MPa·m1/2, respectively. The at room temperature and 650℃ are
0.8MPa·m1/2 and 11.1MPa·m1/2, respectively. It can be found that the increase of
temperature increases the intrinsic and extrinsic contributions to fracture
toughness, which is consistent with the analysis results of the fracture features and the
crack propagation path at room temperature and 650℃. It is further confirmed that the
increase of temperature leads to the increase of the inherent property of the material
and the crack propagation tortuosity.
In addition, the intrinsic contribution accounted for 97.5% of the total
fracture toughness while the extrinsic contribution only accounted for 2.5%
at room temperature. It can be found that the is much higher than the ,
indicating that the fracture toughness at room temperature is dominated by the
intrinsic contribution, In other words, the fracture toughness at room temperature is
dominated by the inherent property of the material. However, the accounted for
83.0% of the total fracture toughness and the accounted for 17% at 650℃.
It can be found that the extrinsic contribution to fracture toughness at high
temperature is significantly increased compared with that at room temperature,
indicating that the fracture toughness at high temperature is dominated by both the
intrinsic and extrinsic contributions. In other words, the fracture toughness at 650℃ is
dominated by the inherent property of the material and the crack propagation
tortuosity. The above analysis results are consistent with that of fracture mechanism at
room temperature and 650℃.
Although the percentage of the extrinsic contribution to fracture toughness has
increased at 650℃, it can still be found that the intrinsic contribution is the most
important contribution to the total fracture toughness, that is, the inherent property of
the material is the most important factor to maintain high fracture toughness.
3.4.2 Crack tip plastic zone model
It has been found from the section 3.4.1 that the intrinsic contribution is the most
important contribution to the total fracture toughness. The crack tip plastic zone, as a
primary factor affecting the intrinsic contribution to fracture toughness, is related to
the inherent property of the material [39]. Therefore, the crack tip plastic zone will be
discussed in this section.
Based on the previous simplified shape of the crack tip plastic zone, a fracture
toughness model containing the real height of the crack tip plastic zone l has been
established, as shown in Equation (8). Equation (8) can be changed into the following
form.
( )
= $
(10)
!" #

Therefore, a crack tip plastic zone model can be established to calculate the real
height of the crack tip plastic zone l. In Equation (10), the fracture toughness KIC can
be obtained by the experiment, the crack propagation tortuosity R can be obtained by
Equation (1), the Young's modulus E can be obtained by the elastic deformation stage
during the tensile process, and the specific energy of deformation Av can be replaced
with the deformation work at uniform deformation stage during the tensile process.
The specific energy of deformation Av can be given as
*
%& = ', + () (11)

Where εu is the strain at uniform deformation stage during the tensile process.
Fig.8 shows the tensile curves of Ti-22Al-25Nb alloy with bimodal size lamellar
O phase microstructures at room temperature and 650℃. It can be observed that the
stress at room temperature does not decrease during the process of tensile test,
indicating that there is no unsteady flow during the whole plastic deformation process.
However, when the strain is about 0.1, the stress at 650℃ begins to decrease gradually.
Av and E can be obtained by the tensile curves. The results show that the Av are
120.8MPa and 63.2MPa at room temperature and 650℃ respectively, and the E are
122.6GPa and 27.4GPa at room temperature and 650℃ respectively. The Poisson's
ratio ν is taken as 0.3. Finally, the real height of the crack tip plastic zone l at room
temperature and 650℃ is calculated by Equation (11), as shown in Table 2.
 Fig. 8. The tensile curves of Ti-22Al-25Nb alloy with bimodal size lamellar O phase
microstructures at room temperature and 650℃
As shown in Table 2, the l values at room temperature and 650℃ are 0.08mm
and 1.97mm, respectively. It can be found that the real height of the crack tip plastic
zone l increases with the increase of temperature and it is consistent with the tendency
of KIC. With the increase of temperature, the volume of the crack tip plastic zone
increases remarkably, and it would consume and absorb the more crack tip plastic
deformation work, which causes the bluntness of the crack tip. Therefore, the crack
propagation resistance at 650℃ is greater than that at room temperature, which leads
to a significant increase in fracture toughness. Generally, the existence of the crack tip
plastic zone results in the bluntness of the crack, which will leave the traces of plastic
deformation on the macro fracture surface of the CT samples. In terms of the macro
fracture surface, the macro fracture surface at room temperature shows the features of
abundant cleavage facets, which indicate that there is no obvious plastic deformation
trace on the macro fracture surface, so the real height of the crack tip plastic zone l is
very low. However, the macro fracture surface at 650℃ is rough and uneven, and the
shear lip zone can be clearly observed. These features indicate the macro fracture
surface has a lot of plastic deformation traces, so the real height of the crack tip
plastic zone l is very high. Furthermore, it can be seen that the analysis results of the
crack tip plastic zone model are consistent with that of the macro fracture. It is worth
noting that, based on the simplified principle of the crack tip plastic zone, the real
height of the crack tip plastic zone is proportional to the volume of the crack tip
plastic zone, that is, the larger the height of the crack tip plastic zone is, the larger the
volume of the crack tip plastic zone is. Therefore, the crack tip plastic zone model can
be used to qualitatively compare the size of the crack tip plastic zone.

Table 2 The real height l of the crack tip plastic zone at room temperature and 650℃

T(℃) (MPa·m1/2) R Av(MPa) E(GPa) ν l( mm)

20 32.5 1.08 120.8 122.6 0.08

0.3
650 65.4 1.45 63.2 27.4 1.97

In addition, the real height of the crack tip plastic zone l calculated by the crack
tip plastic zone model (Equation (10)) is compared with the l calculated by the
theoretical model and the finite element model. The yield strength at room
temperature and 650℃ obtained by the tensile curves in Fig.8 are 912MPa and
763MPa, respectively.
Table 3 shows the l values calculated by three crack tip plastic zone (CTPZ)
models. The real height of the crack tip plastic zone at 650℃ is greater than that at
room temperature for the three CTPZ models, which again proves that the increase of
temperature improves the size of the crack tip plastic zone. Moreover, the l values
calculated by three CTPZ models are 0.08mm, 0.30mm and 0.38mm at room
temperature, indicating that the l value calculated by Equation (10) is quite different
from that calculated by the latter two models. However, the l values calculated by
three CTPZ models are 1.97mm, 1.78mm and 2.22mm at 650℃. It can be found that
the l values are in good agreement. The crack tip plastic zone model established in
this paper is a physical model that considers the KIC, E, R and Av, while the theoretical
model and finite element model are both empirical models that only considers the KIC
and . Besides, in terms of the three CTPZ models, the KIC of the material is larger,
the real height of the crack tip plastic zone l is larger, and the error between Equation
(10) and the latter two models will decrease. Ti2AlNb based alloy, as an intermetallic
compound, usually has intrinsic brittleness, and the fracture mode at room
temperature is the cleavage fracture, which indicates that the material at room
temperature has low toughness, resulting in large error between Equation (10) and the
latter two models. However, the fracture mode at high temperature is the ductile
fracture, which indicates that the toughness of the material is significantly improved,
resulting in small error between Equation (10) and the latter two models. Therefore, to
some extent, it can be concluded that the crack tip plastic zone model established in
this paper can be used to quantitatively calculate the height of the crack tip plastic
zone.

Table 3 The real height of the crack tip plastic zone l calculated by the three CTPZ
models
The theoretical The finite element
The CTPZ model Equation (10)
model model

4(1 − . $ ) 0.24 $
0.3 $
= $
= =
/0%&
Equation $ $

l(mm) (RT) 0.08 0.30 0.38

l(mm) (650℃) 1.97 1.78 2.22

4. Conclusions
1) The fracture toughness KIC of Ti-22Al-25Nb alloy with bimodal size lamellar O
phase microstructures at room temperature and 650℃ are 32.5 MPa·m1/2 and 65.4
MPa·m1/2, respectively. The fracture mechanism at room temperature is the
quasi-cleavage fracture while that at high temperature is dimple fracture. Therefore,
the fracture toughness KIC of Ti-22Al-25Nb alloy is very sensitive to temperature.
2) The fracture mechanism of Ti-22Al-25Nb alloy with bimodal size lamellar O phase
microstructures changes from brittle fracture at room temperature to ductile fracture at
650℃, which indicates that the inherent property of the material (intrinsic
contribution) is improved. Meanwhile, the crack propagation path becomes more
tortuous with the increase of temperature, which indicates that the crack propagation
tortuosity (extrinsic contribution) is also improved. Therefore, the fracture toughness
at 650℃ is significantly improved.
3) Through simplifying the shape of the crack tip plastic zone, a fracture toughness
model is established to quantitatively calculate the intrinsic and extrinsic contributions
to fracture toughness. The intrinsic contribution increased from 31.7MPa·m1/2 at room
temperature to 54.3MPa·m1/2 at 650℃, the extrinsic contribution increased from
0.8MPa·m1/2 at room temperature to 11.1MPa·m1/2 at 650℃. Moreover, the fracture
toughness is dominated by intrinsic contribution at room temperature while the
fracture toughness is dominated by both the intrinsic and extrinsic contributions at
650℃. Besides, the intrinsic contribution is the most important factor to the total
fracture toughness.
4) A crack tip plastic zone model is established to calculate the real height of the crack
tip plastic zone. The height of the crack tip plastic zone l are 0.08mm and 1.97mm at
room temperature and 650℃, respectively, which indicates that the size of the crack
tip plastic zone significantly increases with the increase of temperature, thereby
leading to a significant increase in fracture toughness. Moreover, to some extent, the
crack tip plastic zone model established in this paper can be used to quantitatively
calculate the height of the crack tip plastic zone.
Acknowledgements
The authors would like to acknowledge the financial supports from the Program
of National Key Research and Development Plan of China (NO. 2016YFB0301203).
In addition, this work was supported by China Postdoctoral Science Foundation
(2019M653727), Natural Science Foundation of Shaanxi province of China
(2020JQ-156) and the Fundamental Research Funds for the Central Universities
(3102019TS0404).
Data availability
The raw/processed data required to reproduce these findings cannot be shared at
this time as the data also forms part of an ongoing study.
References
[1] D. Banerjee, A.K. Gogia, T.K. Nandi, V.A. Joshi, A new ordered orthorhombic
phase in a Ti3Al-Nb alloy, Acta Metall. 36 (1988) 871-882. DOI:
10.1016/0001-6160(88)90141-1.
[2] D. Banerjee, The intermetallic Ti2AlNb, Prog. Mater. Sci. 42 (1997) 135-158.
DOI: 10.1016/S0079-6425(97)00012-1.
[3] J. Kumpfert, Intermetallic alloys based on orthorhombic titanium aluminide, Adv.
Eng. Mater. 3(2001) 851-864. DOI: 10.1002/1527-2648(200111)3:113.0.CO;2-G.
[4] C.J. Boehlert, B.S. Majumdar, V. Seetharaman, D.B. Miracle, Part I: The
microstructural evolution in Ti-Al-Nb O+Bcc orthorhombic alloys, Metall. Mater.
Trans. A 30(1999) 2305-2323. DOI: 10.1007/s11661-999-0240-4.
[5] R.G. Rowe, D.G. Konitzer, A.P. Woodfield, Tensile and creep behavior of ordered
orthorhombic Ti2AlNb-Based alloys, MRS Online Proc. Libr. 1990 213. DOI:
10.1557/PROC-213-703.
[6] A. Partridge, E.F.J. Shelton, Processing and mechanical property studies of
orthorhombic titanium-aluminide-based alloys, Air and Space Europe. 3(2001)
170-173. DOI: 10.1016/S1290-0958(01)90085-1.
[7] J.K. Zhang, X.Q. Cheng, Z.N. Li, Total fatigue life prediction for Ti-alloys
airframe structure based on durability and damage-tolerant design concept, Mater.
Des. 31(2010) 4329-4335. DOI: 10.1016/j.matdes.2010.03.052..
[8] S.H. Nahm, C.M. Suh, M.W. Jung, Application of damage tolerance approach for
turbine disk life extension, Int. J. Mod. Phys. B 17(2003) 1916-1921. DOI:
10.1142/S0217979203019873.
[9] Y.Q. Zhao, H.L. Qu, Research on high strength and high toughness titanium alloy
with damage tolerance, Mater. Sci. Forum 654-656(2010) 586-589. DOI:
10.4028/www.scientific.net/msf.654-656.586.
[10] A.K. Gogia. High-temperature titanium alloys, Defence Sci. J.,55(2005) 49-173.
DOI: 10.14429/dsj.55.1979.
[11] S.J. Eck, A.J. Ardell, Fracture toughness of Ti-46.5Al-2.1Cr-3.0Nb-0.2W from
finite element analysis of miniaturized disk-bend test results, Intermetallics 6
 (1998) 471-477. DOI: 10.1016/s0966-9795(97)00047-2.
[12] Y. Chen, H. Niu, F. Kong, S. Xiao, Microstructure and fracture toughness of a β
phase containing TiAl alloy, Intermetallics 19 (2011) 1405-1410. DOI:
10.1016/j.intermet.2011.05.006.
[13] L. Yu, X. Song, M. Zhang, Z. Jiao, H. Yu, High-temperature fracture toughness of
high Nb-containing TiAl alloys, Chinese Journal of Engineering, 38(2016)
379-384. DOI: 10.13374/j.issn2095-9389.2016.03.012.
[14] X. Shi, W. Zeng, Q. Zhao, The effects of lamellar features on the fracture
toughness of Ti-17 titanium alloy, Mater. Sci. Eng. A 636 (2015) 543-550. DOI:
10.1016/j.msea.2015.04.021.
[15] Y. Zheng, W. Zeng, D. Li, Fracture toughness of the bimodal size lamellar O
phase microstructures in Ti-22Al-25Nb (at.%) orthorhombic alloy, J. Alloys
Compd. 709 (2017) 511-518. DOI: 10.1016/j.jallcom.2017.03.184.
[16] D. Eylon, S. Fujishiro, P.J. Postans, High-temperature titanium alloys—a review,
JOM. 36(1984) 55-62. DOI: 10.1007/BF03338617.
[17] X.G. Fan, H.J. Zheng, P.F. Gao, Phase fraction evolution in hot working of a
two-phase titanium alloy: experiment and modeling, Rare Met. 2017,
36(010):769-779. DOI: 10.1007/s12598-017-0950-5.
[18] S. Emura, A. Araoka, M. Hagiwara, B2 grain size refinement and its effect on
room temperature tensile properties of a Ti-22Al27Nb orthorhombic intermetallic
alloy, Scr. Mater. 48 (2003) 629-634. DOI: 10.1016/s1359-6462(02)00462-1.
[19] W. Wang, W. Zeng, C. Xue, X. Liang, J. Zhang, Designed bimodal size lamellar
O microstructures in Ti2AlNb based alloy: Microstructural evolution, tensile and
creep properties, Mater. Sci. Eng. A 618 (2014)288-294. DOI:
10.1016/j.msea.2014.09.035.
[20] Y. Zheng, W. Zeng, D. Li, Effect of orthorhombic case on the creep rupture of
Ti-22Al-25Nb (at%) orthorhombic alloy, Mater. Sci. Eng. A 696 (2017) 529-535.
DOI: 10.1016/j.msea.2017.05.003.
[21] R.O. Ritchie, Mechanisms of fatigue-crack propagation in ductile and brittle
solids, Int. J. Fract. 100 (1999) 55-83. DOI: 10.1023/a:1018655917051.
[22] M. E. Launey, R. O. Ritchie, On the fracture toughness of advanced materials,
Adv. Mater. 2010, 21(21):2103-2110. DOI: 10.1002/adma.200803322.
[23] G. Lütjering, J. Albrecht, C. Sauer, T. Krull, The influence of soft, precipitate-free
zones at grain boundaries in Ti and Al alloys on their fatigue and fracture
behavior, Mater. Sci. Eng. A 468 (2007)201-209. DOI:
10.1016/j.msea.2006.07.168.
[24] X. Shi, W. Zeng, C. Shi, H. Wang, Z. Jia, The fracture toughness and its
prediction model for Ti–5Al–5Mo–5V–1Cr–1Fe titanium alloy with
basket-weave microstructure, J. Alloys Compd. 632 (2015) 748-755. DOI:
10.1016/j.jallco m.2 015.01.217.
[25] D. Eylon, J. A. Hall, C. M. Pierce, Microstructure and mechanical properties
relationships in the Ti-11 alloy at room and elevated temperatures, Metall. Trans.
A 7 (1976) 1817-1826. DOI: 10.1007/BF02659811
[26] A.S. Argon, Y. Qiao, Cleavage cracking resistance of large-angle grain
boundaries in Fe-3wt%Si alloy, Philos. Mag. A 82 (2002) 3333-3347. DOI:
10.1080 /014 18610208240445.
[27] Y. Qiao, A. S. Argon, Cleavage crack-growth-resistance of grain boundaries in
polycrystalline Fe–2%Si alloy: experiments and modeling, Mech. Mater. 35
(2003) 129-154. DOI: 10.1016/s0167-6636(02)00194-1.
[28] ASM Hand Book, Volume 12: Fractography, ninth ed., ASM International, 1987,
pp. 40-43.
[29] C. J. Boehlert, Part III. The tensile behavior of Ti-Al-Nb O+Bcc orthorhombic
alloys, Metal. Mater. Tran. A 32(2001)1977-1988. DOI:
10.1007/s11661-001-0010-4
[30] M. Hagiwara, S. Emura, A. Araoka, The effect of lamellar morphology on tensile
and high-cycle fatigue behavior of orthorhombic Ti-22Al-27Nb alloy, Metal.
Mater. Tran. A, 35 (2004) 2161-2170. DOI: 10.1007/s11661-004-0164-y
[31] X. Jiang, Fractal analysis of the anisotropic fracture toughness of materials, J.
Mater. Sci. Lett. 11 (1992) 1379-1380. DOI: 10.1007/bf00729368.
[32] J. Małecka, Investigation of the oxidation behavior of orthorhombic Ti2AlNb
 Alloy. J. Mater. Eng. Perform. 24 (2015) 1834-1840.
DOI:10.1007/s11665-015-1449-6.
[33] F. Popille, J. Douin, Comparison of the deformation microstructures at room
temperature in O and B2 phases of a Ti2AlNb alloy, J. de Physique, 6 (1996)
C2-211-216. DOI: 10.1051/jp4:1996229.
[34] F. Popille, J. Douin, The dislocation microstructure in orthorhombic O Ti2AlNb
deformed between room temperature and 800℃, Philos. Mag. A 73 (1996)
1401-1418. DOI: 10.1080/01418619608245141.
[35] B. Shao, D. Shan, B. Guo, Plastic deformation mechanism and interaction of B2,
α2, and O phases in Ti-22Al-25Nb alloy at room temperature, Int. J. Plasticity 113
(2019) 18-34. https://doi.org/10.1016/j.ijplas.2018.09.004.
[36] Y.I. Ragozin, Y.Y. Antonov, Method of accelerated fracture toughness KIC testing
of metallic materials, Strength Mater. 16 (1984) 179-184. DOI: 10.1007/BF01530
053.
[37] N.A. Makhutov, Y.C. Matvienko, Griffith theory and development of fracture
mechanics criteria. Mater. Sci. 29 (1993) 316-319. DOI: 10.1007/BF00558976.
[38] Q.P. Zhong, Fractography. Beijing: Higher education press. 2006. pp. 30-33.
[39] R.C. Jia, W.D. Zeng, S.T. He, The analysis of fracture toughness and fracture
mechanism of Ti60 alloy under different temperatures, J. Alloys Compd. 810
(2019). DOI: 10.1016/j.jallcom.2019.151899.