See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315871007

High-Temperature Oxidation Kinetics and Behavior of Ti–6Al–4V Alloy

Article in Oxidation of Metals · December 2017

DOI: 10.1007/s11085-017-9770-0

CITATIONS READS

74 2,329

5 authors, including:

Yu Wei Qingwu Cai

University of Science and Technology Beijing 31 PUBLICATIONS 443 CITATIONS
142 PUBLICATIONS 1,055 CITATIONS
SEE PROFILE
SEE PROFILE

Cheng Lei
University of Science and Technology Beijing
19 PUBLICATIONS 180 CITATIONS

SEE PROFILE

All content following this page was uploaded by Cheng Lei on 18 February 2020.

The user has requested enhancement of the downloaded file.

Oxid Met
DOI 10.1007/s11085-017-9770-0

ORIGINAL PAPER

High-Temperature Oxidation Kinetics and Behavior

of Ti–6Al–4V Alloy

Entao Dong1,2 • Wei Yu1 • Qingwu Cai1,2 •

Lei Cheng1 • Jiaxin Shi1

Received: 14 January 2017 / Revised: 21 February 2017

Ó Springer Science+Business Media New York 2017

Abstract Owing to the high-temperature reactivity of titanium, the oxidation and

alloying of titanium during hot working processes is an important variable. The
oxidation behavior of Ti–6Al–4V alloy in air was investigated at various temper-
atures between 850 and 1100 °C for different times. The oxidation kinetics were
determined by isothermal oxidation weight gain experiments. The results showed
that the oxidation kinetics approximately obeyed a parabolic law. The activation
energy of oxidation was estimated to be 199 and 281 kJ mol-1 when temperature
was above and below the beta transformation temperature (Tb), respectively. A
model to predict oxidation extent was established based on experimental observa-
tions. The oxide scales mainly consisted of TiO2 with a small amount of Al2O3 and
TiVO4. The alpha case was defined as solid solution formed because of oxygen
diffusion into the substrate. The difference in the morphology and the formation
mechanism of the alpha case at different temperature ranges was mainly owing to
the participation of the grain boundary and grain orientation of the nucleation site.

Keywords Ti–6Al–4V alloy  Oxidation kinetics  Oxidation scale  Alpha case 

Formation mechanism

Introduction

Titanium and its alloys are very well known because of their desirable mechanical
and chemical properties. They have received much attention recently owing to their
wide range of applications in the aerospace, biomedical, and chemical industries

& Wei Yu
[email protected]; [email protected]
1
University of Science and Technology Beijing, 30 Xueyuan Road, Beijing 100083, China
2
Collaborative Innovation Center of Steel Technology, 30 Xueyuan Road, Beijing 100083,
China

123
 Oxid Met

[1–4]. The most common titanium alloy is Ti–6Al–4V, which contains 6% of the a-
stabilizing element Al and 4% of the b-stabilizing element V. Ti–6Al–4V alloy is an
a ? b titanium alloy and accounts for more than 50% of the titanium alloy
production all over the world [5]. Aluminum and vanadium are beneficial because
they substantially reduce the amount of oxygen dissolved on the surface of Ti–6Al–
4V compared with unalloyed titanium [6]. However, the use of titanium alloys is
often limited to temperatures below 600 °C because of their oxidation behavior and
the diffusion of oxygen. The properties of titanium alloys degenerate remarkably
and have limited resistance to hot air as the temperature increases. It is common for
metals oxidized at high temperatures, which greatly influences their properties,
surface finish, and service life [7]. The high affinity of titanium to oxygen causes the
formation of a very thin oxide layer on the surface even in air at room temperature.
Furthermore, the oxidation of titanium alloys in air at high temperature results not
only in oxidation scales but also in solid solution hardening of the surface because
of oxygen diffusion, leading to the formation of a hardened zone known as an
‘‘alpha case’’ [8]. Therefore, it is necessary to remove any alpha case from titanium
alloys by mechanical or chemical methods [9].
Recently, many studies have been devoted to the oxidation behavior of titanium
alloys under different conditions [10–17]. Du et al. [10] found that multilayered
oxide scales with alternating layers of Al2O3 and TiO2 developed on Ti–6Al–4V
oxidized at 750 °C in air. Kumar et al. [11] investigated the oxidation behavior of
Ti–6Al–4V alloy at 500, 650, and 800 °C and concluded that there was no direct
correlation between the thickness of the obtained oxide layer as a function of
temperature and the time employed for oxidation and the corrosion protection
ability. Poquillon et al. [12] established a correlation between the hardness and the
oxygen content in the oxygen-affected zone, which indicated that the oxygen
content was constant at a given depth in both the a and b phases. Frangini et al. [13]
studied the air oxidation behavior of Ti–6Al–4V alloy between 600 and 700 °C and
indicated the onset of a duplex nature starting from 650 °C, which became more and
more pronounced as a function of time and temperature. Also, many methods have
been used to modify the surface properties of Ti-based alloys in order to improve
oxidation resistance, such as surfaces treatment with coatings [18, 19].
The hot working of Ti–6Al–4V includes billet heating, forging, rolling, and
extruding. Heat treatment is also needed to obtain good performance for the
application in titanium alloy production. Although there have been numerous
investigations about the oxidation behavior of Ti–6Al–4V alloy from an application
point of view [20], very few attempts have been made to correlate the oxidation
mechanism at temperatures above 850 °C. The present study is an attempt to
elaborate the evolution of the oxidation kinetics and the oxides developed of the Ti–
6Al–4V alloy and establish the formation mechanism of alpha case for short
oxidation durations at high temperatures near Tb in air. The selected oxidation
treatment conditions are of interest in the hot working temperature including billet
heating, hot forging, rolling, and heat treatment temperature. The oxides is often
avoided by using high-temperature coatings, or removed by machining or chemical
milling. The necessity of using these additional processes is one of the reasons for
higher manufacturing cost of titanium alloys. The effect of temperature and time on

123
Oxid Met

the oxidation kinetics and oxidation scale formation in the hot working temperature
range of 850–1100 °C is investigated in order to provide reasonable process
parameters for improving the oxidation resistance. A quantitative model of
oxidation weight gain would also be helpful for this purpose. Therefore, decreasing
the cost and understanding the oxidation theory of Ti–6Al–4V alloy is urgently
needed.

Experimental Procedures

The material used in the experiment was a commercial Ti–6Al–4V alloy obtained
by forging and machining. The initial microstructure consisting of 20 vol%
transformed b phase and 80 vol% equi-axed primary a phase is shown in Fig. 1. Tb
was measured to be approximately 980 °C by thermal dilatation method. The
material was cut into specimens with a size of 10 mm 9 10 mm 9 5 mm. The
surface of all the specimens was polished with 200–1500 mesh SiC abrasive paper.
The specimens were then ultrasonically cleaned in acetone and alcohol and dried in
hot air.
The isothermal oxidation tests were carried out at temperatures between 850 and
1100 °C for time between 0.5 and 3 h using an air muffle furnace in normal
atmosphere. The specimens were placed in a corundum crucible and quickly put
into the furnace heating zone after the required temperature was stable. The
specimens were taken out from the muffle furnace after the appropriate oxidation
time and then cooled down to ambient temperature in air. The specimens oxidized in
a corundum crucible were dried at 70 °C for 60 min in a thermotank. The weight of
the specimens before and after the oxidation was measured by an electronic balance
(AL204-IC Mode) with an accuracy of ±10-4 g. After grinding and polishing, the
microstructure was exposed by etching in Kroll’s reagent (5 vol% HF ? 10 vol%
HNO3 ? 85 vol% H2O).
Metallographic analysis was conducted by optical microscopy (OM-Zeiss
Axiovert 40) and scanning electron microscopy (SEM-Zeiss Ultra 55). Energy-

Fig. 1 Microstructure of Ti–

6Al–4V alloy specimen

123
 Oxid Met

dispersive X-ray spectroscope (EDS) was used to determine the chemical

compositions and element profiles. X-ray diffraction (XRD-Bruker D8) with a Cu
Ka radiation was utilized for identification of the phase of the surface oxides, and
the diffraction angle range was between 10° and 90° with a step increment of 0.02°.

Results

Oxidation Kinetics

Figure 2 shows the isothermal weight gain of Ti–6Al–4V alloy as a function of

oxidation time at temperature between 850 and 1100 °C. The weight gain curves in
Fig. 2 show that the alloy examined has obvious different oxidation behaviors at
different temperatures and times. As the oxidation temperature increased above Tb,
the weight gain curves were significantly higher than that below Tb. The weight gain
increased rapidly in the initial oxidation within 90 min, and then, the weight change
gradually slow down and was relatively stable with increasing time.
In general, the high-temperature oxidation kinetics can be described by the
following equation:
 
DW n
¼ kp t ð1Þ
A
where DW is the weight gain, A is the surface area of the specimen, kp is the
oxidation rate constant, t is the oxidation time, and n is the rate exponent. To
determine the value of n, the slope of the linear fit of ln DW versus ln t was
calculated from the weight gain. The result listed in Table 1 indicated that the value
of n for Ti–6Al–4V alloy was approximately 2, which roughly yielded parabolic
oxidation kinetics. The oxidation rate constant kp was calculated from the plot of
(DW/A)2 versus time, and the results are displayed in Table 1.

Fig. 2 Oxidation kinetics

curves of Ti–6Al–4V alloy in 50 1100 °C
the range of 850–1050 °C 1050 °C
1000 °C
Weight gain (mg⋅cm -2 )

40 950 °C
900 °C
850 °C
30

20

10

0
0.5 1.0 1.5 2.0 2.5 3.0
Time (h)

123
Oxid Met

Figure 3 shows the good linear fit between ln kp and 1/T plot. The activation
energy Q of oxidation above and below Tb obtained from the plot was calculated to
be 199 and 281 kJ mol-1, respectively. It is well known that the diffusion ability of
oxygen in b-Ti with a body-centered cubic structure is greater than that of oxygen in
a-Ti with a hexagonal close-packed structure. This is consistent with the decrease in
the oxidation activation energy as temperature exceeds Tb.
In this study, the analysis of the extent of oxidation was based on various
oxidation temperatures and times. Therefore, the temperature-compensated time
parameter has a unit of time, denoted as:
 
Q
h ¼ t exp  ð2Þ
RT
Combining Eqs. (2), (1) can be transformed into the follow form:
 
DW
ln ¼ 0:5 ln h þ 12:03 ðT [ Tb Þ
A
  ð3Þ
DW
ln ¼ 0:5 ln h þ 16:12 ðT\Tb Þ
A
According to Fig. 2 and Eq. (2), the fit line of ln DW/A versus ln h is obtained based
on the measured weight gains. As shown in Fig. 4, the predicted values are always
slightly higher than the measured values because the oxygen absorbed by Ti–6Al–
4V alloy during oxidation is used not only for the formation of oxide scale but also
for the formation of the alpha case [21]. This is the reason for the slight difference
between the predicted and measured values.

Phase Identification

Figure 5 presents the XRD patterns of the specimens oxidized under various
conditions. The oxidized surface of the alloy consisted mainly of rutile-type TiO2
and a small amount of Al2O3 and TiVO4 (a solid solution of TiO2 and VO2). In
contrast to the previous reports [22, 23], there were no Ti diffraction peaks in the
XRD pattern because the scale was very thick after severe oxidation and the X-rays
could not penetrate the oxides to detect the Ti substrate. As the temperature
increased, the diffraction peaks intensity and the rutile peaks (TiO2 and TiVO4)
became more dominant with higher intensity, as shown in Fig. 5b. Furthermore,
only the intensity of the diffraction peaks showed tiny difference after oxidation for
different time periods at the same temperature.

Table 1 Rate exponent n and rate constant kp of Ti–6Al–4V alloy oxidized at 850–1100 °C
Temperature/°C 1100 1050 1000 950 900 850

n 1.7 2.2 1.8 1.5 1.1 1.7

kp/mg2 cm-4 h-1 816 324 206 85 43 7

123
 Oxid Met

Fig. 3 Arrhenius plot of 8

parabolic rate constant kp for
oxidation of Ti–6Al–4V alloy in
Q=199 kJ⋅mol-1
the range of 850–1100 °C

ln k p/mg 2⋅cm-4 ⋅h -1
6

Q =281 kJ⋅mol -1
4

-4 -4 -4 -4 -4
7.0x10 7.5x10 8.0x10 8.5x10 9.0x10
-1 -1
T /K

4 a 3 b
Predicted
Predicted ln(Δ W/A)= 0.5lnθ +16.12
ln(ΔW/Α)=0.5lnθ +12.03
ln(ΔW/A)
ln (ΔW/A)

2
3

Measured 1
Measured
ln (Δ W/Α)=0.51lnθ +12.19 ln (ΔW/Α)= 0.56 lnθ +17.83
2
0
-20 -19 -18 -17 -16 -31 -30 -29 -28 -27 -26
lnθ lnθ

Fig. 4 Comparison between the predicted and measured plot of ln DW/A versus ln h a T [ Tb, b T \ Tb

a ο ο −TiVO4 ο− TiVO4
♦
♦ − Rutile (TiO2)
b ο
♦ ♦ −Rutile(TiO2)
ο ◊− Al2 O3 ◊− Al 2O 3
♦
ο
Intensity/(a.u.)

Intensity/(a.u.)

♦
ο
♦
ο ο ο ο ο
♦ ♦ ο ♦ οο ♦ ♦ ο
◊ ◊♦♦ ♦
ο ♦ ◊ ◊ ◊ ♦ ◊ ◊♦♦ οο ο
♦ ♦♦ ♦
◊ 3h
◊ ◊ ◊
◊ ◊
2h 2h
1h 1h
20 30 40 50 60 70 80 90 20 30 40 50 60 70 80 90
2θ/(°) 2θ/(°)

Fig. 5 XRD patterns of surface of Ti–6Al–4V alloy oxidized under various conditions a 900 °C for 1,
2 h; b 1050 °C for 1, 2, 3 h

At temperatures above 915 °C, the TiO2 crystal has a tetragonal structure (or
called rutile structure): lattice constant a = 4.584 Å, c = 2.953 Å, Z = 2. VO2 also
has a rutile structure, which can dissolve in the rutile TiO2. That is why there is a

123
Oxid Met

small amount of TiVO4 in the base of the rutile-type TiO2 scales, but not individual
VO2 crystal. It has been observed that, although the solid solution stabilizes the
reduced valence state of vanadium (V4?), oxygen diffusion in the rutile phase may
lead to localized oxidation of V4? with consequent formation of either anion
vacancies or interstitial cations [13]. However, a-alumina has a rhombohedra
structure and its lattice constants are a = 4.75 Å, c = 12.97 Å, which are quite
different from that of the rutile structure. Therefore, the alumina can only nucleate
and grow independently by the diffusion aluminum atoms from the substrate to the
scale surface.

Surface Morphology of Oxide Scale

Figure 6 shows the surface morphologies of specimens oxidized for 2 h at 850, 950,
1000, and 1100 °C. At a low temperature 850 °C, the oxide scales have fine grains
with directionality, but they are not uniform, as shown in Fig. 7a. The directionality
may be related to grinding deformation on the specimen surface. Oxide grains with
a fine size tend to congregate along grinding marks, leading to the formation of an
undulating oxide layer on the surface. As the temperature elevated, the surface
oxide scale crystals grow and the surface becomes flat and smooth.
Figure 7 presents the typical morphologies and EDS analysis results of the
surface oxide scale at 950 °C. As shown in Fig. 7a, b, concaves and convexes were
observed on the oxides surface at the initial oxidation stage. As the oxidation time

Fig. 6 SEM micrographs of surface of specimens oxidized for 2 h at a 850 °C, b 950 °C, c 1000 °C, and
d 1100 °C

123
 Oxid Met

a b

c d

e
f 1 2 3
Element Weight/%
O 50.07 38.88 50.88
Al 40.83 9.32 41.87
Ti 16.18 50.03 6.78
V 0.92 1.86 0.47

Fig. 7 SEM micrographs and EDS results of oxide scales on the surface oxidized at 950 °C for (a,
b) 0.5 h, (b, c, d) 3 h, and (f) EDS analysis at positions marked

increased, the morphology of the partial surface became denser, as shown in

Fig. 7d, e. The EDS spectra of the surface oxides shown in Fig. 7f confirmed that
the polyhedron-type structure was mainly composed of Al-oxide (Al2O3) at position
1 and 3, and the fine rods on the convex regions mainly consisted of Ti-oxide
(TiO2). The EDS analysis revealed that the surface oxides were mostly covered with
Al2O3 and TiO2 after oxidation for 30 min. The wavy surface increased the surface
area for oxygen absorption and accelerated the oxidation. The continuous outward
growth of polyhedron-type Al2O3 is evident by their ends joining together over the
surface after oxidation for 3 h, as shown in Fig. 7e.
Figure 8 shows the surface morphologies of oxides obtained at various durations
at 1000 °C. When temperature was above Tb, in addition to the formation of fine
polyhedron and rod-type structures, dense massive structures were formed, as
shown in Fig. 8. With increased oxidation time, some of the crystals grew outward
and changed from having a fine acicular shape to a thick and long rod shape, as

123
Oxid Met

a b

c
d 1 2 3
Element Weight/%
O 32.79 50.26 52.24
Al 0.70 2.98 40.97
Ti 64.38 45.26 6.48
V 2.13 1.5 0.31

Fig. 8 SEM micrographs and EDS results of oxide scales on the surface oxidized at 1000 °C for a 0.5 h,
b, c 3 h, and f EDS analysis at the positions marked

shown in Fig. 8a, b. The EDS analysis clearly revealed that the dense massive
structure in position 1 was mainly composed of Ti and O, whereas a higher amount
of V is detected as compared with position 2 and 3. According to a previous study,
V could promote crystal transformation of TiO2 at a high temperature [24].
Furthermore, the massive structures had a rutile TiO2 with a small amount of
TiVO4. Similarly, the polyhedron particles and rods were associated with Al2O3 and
TiO2. The results were in accordance with the XRD analysis. Meanwhile, some
cracks and pores appeared in the surface oxides, which may be owing to the
different amounts of shrinkage between the substrate and the oxide scale.

Cross-Sectional Microstructure

Figure 9 presents the cross-sectional microstructure of the specimens oxidized

under different conditions. A brilliant white zone was observed beneath the surface
oxide scales as a result of diffusion of oxygen, which is called alpha case. Figure 9a
shows that only a thin alpha case was formed at 850 °C for 0.5 h. The alpha case
was thicker after longer oxidation time at higher temperatures. This result was
consistent with the oxidation kinetics curves shown in Fig. 2. The shapes of the
grains in the alpha case varied significantly in different temperatures and remained a
similar shape with the phase of the substrate, which had an equi-axed or
Widmanstatten structure. The grains were the shape of fine polygon in the alpha
case when oxidation occurs below Tb (Fig. 9a, b), and coarse lath at oxidation
temperatures above Tb (Fig. 9c, d). The lamellar a phase was not completely

123
 Oxid Met

Fig. 9 Metallographic cross-sectional morphology of Ti–6Al–4V alloy oxidized at a 850 °C for 0.5 h,
b 950 °C for 1.5 h, c 1050 °C for 0.5 h, and d 1050 °C for 2 h

transformed to the b phase in the a ? b temperature range, and the fine a phase still
existed.
The thickness of the alpha case was more than 300 lm when the specimen was
oxidized at 1050 °C for 2 h. Also, it can be found that the grains of alpha case
generated were very irregularly distributed.
The cross-sectional morphologies and EDS line analysis of oxidized specimens
are shown in Fig. 10. The distribution of Ti, O, Al, and V from the surface scales to
substrate was detected. The oxide thickness shown in Fig. 10a, c, and e was between
approximately 40 and 70 lm, depending on the holding time and oxidation
temperature. After oxidation at temperature below Tb, the oxide scale was a
multilayer and partially adhered to the substrate. The oxide scale was thick, dense,
and well adhered to the substrate when the oxidation temperature was above Tb, as
shown in Fig. 10e. Moreover, the concentration of oxygen tended to decrease from
the surface to the substrate. The oxidation products consisted of a thin alumina-rich
outer layer and a titanium oxide and vanadium oxide inner layer. The intensity of
the EDS peaks of titanium and vanadium was almost identical, as shown in
Fig. 10b, d, f, which indicated that the titanium and vanadium oxide formed
simultaneously. The EDS analysis also showed that the alpha case (dark gray color)
had a higher oxygen concentration than the substrate (light gray color), as shown in
Fig. 10c–f. And it is clear that alpha case appeared some cracks on the surface of it
due to its brittleness and the effect of milling in making the SEM sample.

123
Oxid Met

Fig. 10 Cross-sectional morphology and EDS line analysis of oxides for Ti–6Al–4V alloy oxidized at
b EDS line of a 950 °C for 1.5 h; d EDS line of c 1050 °C for 0.5 h; f EDS line of e 1050 °C for 2 h

Discussion

The isothermal weight gain curves in Fig. 2 and the kinetics parameters in Table 1
indicated that the weight gain increase slowed down with prolonged holding time
after a rapid weight gain obeyed a linear law in the first hour. A remarkable increase
in weight was observed above Tb, which basically conformed to a parabolic law in
the entire temperature range. At the temperature above Tb, the rate constant kp
increased significantly compared with that at temperature below Tb. Hence, it is
suggested that oxidation diffusion behavior is governed by a parabolic rate at lower
temperatures. The relatively dense oxide scales prevented further oxidation and
oxygen diffusion into the substrate so that kp was low. Moreover, the oxidation rate
constant above Tb was also affected by some factors such as cracks and pores in the
scales, through which oxygen could enter more easily and quickly reach the
substrate for the formation of an oxide, which resulted in a higher kp. The weight
gain process was controlled by both the surface oxidation rate and diffusion of
oxygen.
Figure 10 shows that the oxidation products consisted of two zones. The high-
intensity peaks of oxygen represented an external zone, in which titanium,
aluminum, and vanadium were diffused and aggregated toward the outer layer. The
lower peaks of oxygen represented inner diffusion of oxygen beneath the scale,
which was probably caused by the dissolution of oxygen into the atomic lattice of
the titanium substrate to form an alpha case. Such solid solution behavior may lead
to lattice distortion of the titanium substrate, resulting in solution strengthening [25].
In a previous report, the oxidation activation energy Q of Ti–6Al–4V alloy was
calculated to be 276 kJ mol-1 at oxidation temperature below Tb [26]. This value is
close to the Q as 281 kJ mol-1 below Tb in this study. The activation energy for the

123
 Oxid Met

diffusion of oxygen in TiO2, titanium in TiO2, and oxygen in Al2O3 was 234, 257,
and 241 kJ mol-1, respectively [27]. In addition, one of the reasons for the low
activation energy value 199 kJ mol-1 above Tb was owing to the formation of
cracks in the oxide, which led to short-cuts in the oxidation pathways [28]. It is also
suggested that the oxidation process was mainly controlled by two mechanisms: the
outward diffusion of titanium and aluminum and the inward diffusion of oxygen
through the oxide scales.
The premise of good oxidation resistance at high temperature is the formation of
oxidation scale to isolate the substrate from the air. However, two other kinds of
stress are formed during the oxidation process: One is growth stress, which is
formed by oxide scale growth during the isothermal oxidation process. The other
one is the thermal stress, which is caused by a difference of thermal expansion or
contraction between the substrate and the oxide scale [29]. In general, the oxide
scales on the alloy surface are in a state of compressive stress. Owing to the
mismatch of the thermal expansion coefficient between the oxide scale and Ti–6Al–
4V alloy substrate, a thermal-to-compressive stress transition is observed when the
alloy is oxidized during cooling in air, which results in severe cracks or exfoliation
of the oxide scale. In addition, exfoliation can occurs through oxide scale folding
under the action of compressive stress (Fig. 7), which is another reason for lower
Q above Tb. As the critical stress of fold increases with the oxide scale thickness,
fold does not occur easily in the thick oxide scale surface.
As shown in Fig. 8, micro-cracks and pores were observed among the rods,
which led to easy oxygen diffusion and hindering of the oxidation resistance.
Preferential growth along certain crystal directions and subsequent growth produce
very fine and long rod-shaped TiO2 crystals after oxidation for a long time or at a
high temperature. The typical composition of Ti–6Al–4V alloy oxidized at high
temperature is shown in Fig. 11.
Figure 9 shows that the thickness of alpha case depends on the oxidation time at
the same temperature. It is suggested that the formation of alpha case was controlled
by the diffusion of oxygen and changed with oxidation time and temperature. It is
known that oxygen has a large solubility in low-temperature hexagonal close-
packed phase (a-Ti) with respect to the high-temperature body-centered cubic phase
(b-Ti). When oxygen diffuses through the oxide scales and dissolves into the Ti
substrate, it can increase the proportion of a phase, because oxygen is the a-Ti
stabilizer that can raise Tb. Owing to the presence of a fine a phase, it acts as a
nucleation site that grows to generate a new a phase instead of direct precipitation.
Although Ti–6Al–4V alloy is composed of a single b phase at temperature above
Tb, the alpha case is mainly a coarse lath shape. This is because oxygen diffuses into
the substrate and transforms the b phase to a ? b phase or an all a phase according
to the oxygen concentration and binary diagram of O-Ti system [30]. The new a
phase is based on the primary b phase and grows along the lamellar structure as
oxygen gradual solution, or is a part of the b phase transformed to lath grain in the
alpha case within the prior b grain colony. In the outer area of the alpha case, alpha
case tends to be polygonal as it is a single phase zone.
As a matter of fact, the irregular grains of alpha case generated stop propagating
eventually until meeting the obstruction of the prior b grain boundaries. The reasons

123
Oxid Met

Fig. 11 Cross-sectional
composition of the oxidation
scale of Ti–6Al–4V alloy
oxidized at high temperature

may be the lamellar structure with different initial directions within the prior b
phase, and alpha case is not easy to cross the different prior b grain boundaries and
continue to grow. Therefore, to study the mechanism of alpha case thickness, the
role of oxygen diffusion as well as the grain boundary and orientation of the
titanium alloy should be considered.

Conclusions

In this work, the isothermal oxidation behavior of Ti–6Al–4V alloy at 850–1050 °C

for 0.5–3 h in air was investigated. The mechanism for the formation of the oxide on
the Ti–6Al–4V alloy was proposed. The following conclusions were drawn:
(1) The oxidation kinetics of the Ti–6Al–4V alloy approximately obeys the
parabolic law. The oxidation activation energy below and above Tb was
determined to be 281 and 199 kJ mol-1, respectively.
(2) The oxidation process was mainly controlled by both the outward diffusion of
titanium, aluminum, and vanadium and the inward diffusion of oxygen
through the oxide scale. The oxide scales of Ti–6Al–4V ally consisted of an
outer layer of dispersive Al2O3 and an inner multilayer of TiO2 and TiVO4.
(3) The oxide scale had good adhesion to the Ti–6Al–4V alloy substrate. No
significant scale spalling was observed at temperatures above 850 °C.
(4) The grain shapes in the alpha case were quite different at different oxidation
temperatures. The grains were fine polygons and coarse lath when the
oxidation temperature was below and above Tb, respectively. This indicated
that the depth of the alpha case was controlled not only by a diffusion process,
but also by the grain boundary and the grain orientation of the nucleation site.

123
 Oxid Met

Acknowledgment The authors would like to acknowledge the financial support from the National Key
Research and Development Program of China (No. 2016YFB0301204).

References
1. R. R. Boyer, Materials Science and Engineering A 213, 103 (1996).
2. I. Gurrappa, Oxidation of Metals 59, 321 (2003).
3. D. Banerjee and J. C. Williams, Acta Materialia 61, 844 (2013).
4. V. Barranco, M. L. Escudero and M. C. Garcı́a-Alonso, Acta Biomaterialia 7, 2716 (2011).
5. P. J. Arrazola, A. Garay, L. M. Iriarte, M. Armendia, S. Marya and F. Le Maı̂tre, Journal of Materials
Processing Technology 209, 2223 (2009).
6. C. Leyens and M. Peters, Titanium and Titanium Alloys, (Wiley-vch, Weinheim, 2003), pp. 187–223.
7. Y. B. Tan, J. L. Duan, L. H. Yang, W. C. Liu, J. W. Zhang and R. P. Liu, Materials Science and
Engineering A 609, 226 (2014).
8. S. Sung and Y. Kim, Journal of Alloys and Compounds 415, 93 (2006).
9. L. Huang, P. Kinnell and P. H. Shipway, Procedia CIRP 6, 594 (2013).
10. H. L. Du, P. K. Datta, D. B. Lewis and J. S. Burnell-Gray, Oxidation of Metals 45, 507 (1996).
11. S. Kumar, T. S. N. Sankara, S. Narayanan, S. Ganesh Sundara Raman and S. K. Seshadri, Materials
Chemistry and Physics 119, 337 (2010).
12. D. Poquillon, C. Armand and J. Huez, Oxidation of Metals 79, 249 (2013).
13. S. Frangini, A. Mignone and F. Dericcardis, Journal Materials Science 29, 1994 (714).
14. T. Izumi, T. Yoshioka, S. Hayashi and T. Narita, Intermetallics 13, 694 (2005).
15. S. Jelliti, C. Richard, D. Retraint, T. Roland, M. Chemkhi and C. Demangel, Surface and Coatings
Technology 224, 82 (2013).
16. N. Somsanith, T. S. N. S. Narayanan, Y. Kim, I. Park, T. Bae and M. Lee, Applied Surface Science
356, 1117 (2015).
17. K. S. Mcreynolds and S. Tamirisakandala, Metallurgical and Materials Transactions A: Physical
Metallurgy and Materials Science 41, 1732 (2011).
18. S. C. Jeng, Surface and Coatings Technology 235, 867 (2013).
19. S. W. Zeng, A. M. Zhao and H. T. Jiang, Applied Surface Science 332, 362 (2015).
20. I. Gurrappa and A. K. Gogia, Surface and Coatings Technology 139, 216 (2001).
21. I. Gurrappa, Journal of Alloys and Compounds 389, 190 (2005).
22. M. M. Arafat, A. S. M. A. Haseeb and S. A. Akbar, Ceramics International 41, 4401 (2015).
23. A. Biswas and J. Dutta, Majumdar. Materials Characterization 60, 513 (2009).
24. H. B. Pan, D. J. Lin, L. X. Liu, S. F. Shen, N. S. Chen and J. L. Huang, Chinese Journal of Materials
Research 15, 535 (2001).
25. H. Guleryuz and H. Cimenoglu, Surface and Coatings Technology 192, 164 (2005).
26. H. L. Du, P. K. Datta, D. B. Lewis and J. S. Burnellgray, Corrosion Science 36, 631 (1994).
27. E. L. Zhang, G. Zeng and S. Y. Zeng, Scripta Materialia 46, 811 (2002).
28. B. G. Velasco and P. B. Aswath, Journal Materials Science 33, 2203 (1998).
29. N. Birks, G. H. Meier and F. S. Pettit, Introduction to the High Temperature Oxidation of Metals,
(Cambridge University Press, Cambridge, 2006), pp. 101–162.
30. J. L. Murray and H. A. Wriedt, Journal of Phase Equilibrium 8, 148 (1987).

123
View publication stats