Materials Science and Engineering A325 (2002) 186193

Brittle-to-ductile transition in multiphase NiAl alloy

C.Y. Cui
a,b,
*, Y.X. Chen
b
, J.T. Guo
a
, Y.H. Qi
a
, H.Q. Ye
b
a
Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016, Peoples Republic of China
b
Laboratory of Atomic Imaging of Solids, Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road,
Shenyang 110016, Peoples Republic of China
Received 20 September 2000; received in revised form 19 April 2001
Abstract
The microstructure of a directionally solidied (DS) NiAl/Cr(Mo,Hf) alloy was studied by scanning electron microscopy (SEM)
and transmission electron microscopy (TEM). This alloy was composed of NiAl, Cr(Mo) and Ni
2
AlHf phases. Tensile testing of
this multiphase NiAl alloy was performed from room temperature to 1323 K at various strain rates from 1.0410
4
s
1
to
1.0410
2
s
1
. It was found that the brittle-to-ductile transition temperature (BDTT) of this alloy was higher than values
reported for many of NiAl-based alloys and dependent on the strain rate. Regardless of strain rate, at temperatures just above
the BDTT, the fracture morphology changed from NiAls cleavage and debonding along the NiAl/Cr(Mo) interface to completely
ductile in nature. It was also found that the apparent activation energy responsible for BDT is 463 kJ mol
1
, which may be
responsible for the higher BDTT. 2002 Elsevier Science B.V. All rights reserved.
Keywords: Microstructure; BDTT; Fracture morphology; The apparent activation energy (Q)
www.elsevier.com/locate/msea
1. Introduction
The intermetallic NiAl has been considered to have a
poor fracture toughness at room temperature, which
makes it unsuitable for high temperature structural
application [1]. The lack of ductility at low temperature
has been attributed to an insufcient number of slip
system due to the operation of 100{hkl} slip, which
provides only three independent slip systems [2,3]. Con-
sequently strain compatibility between grains cannot be
achieved and hence, the room temperature tensile duc-
tility of NiAl is almost nil. Ebrahimi et al. [4,5] have
recently investigated NiAl in single crystal and poly-
crystal forms. In single crystal NiAl, ductility and
toughness are controlled by the formation of cleavage
microcrack on slip bands and their subsequent instabil-
ity [4]. In polycrystalline form, NiAl has a lower-shelf
toughness of 1015 MPa and a BDT temperature
around 673 K [5]. Based on the tensile ductility [6], the
BDT of nominally stoichiometric polycrystalline NiAl
has also been evaluated. Strain rate has a signicant
effect on the BDT with a three order of magnitude
increase in strain rate resulting in a 200 K increase in
the transition temperature for cast and extruded NiAl.
Alloy composition has also a very signicant effect on
the BDTT of NiAl. Small alloying additions or slight
deviations from stoichiometry tend to shift the BDTT
for NiAl to signicantly higher temperature.
Tensile ductility is not a problem after NiAl under-
goes a dramatic brittle-to-ductile transition. While the
mechanisms responsible for this change are not fully
resolved, Ball and Smallman [2] originally suggested
that the brittle-to-ductile transition temperature arise
due to the onset of a diffusion-assisted deformation
process, such as dislocation climb. Miracle [7] has sug-
gested that the brittle-to-ductile transition in NiAl
should be due to additional operation of non-100
dislocation. Noebe et al. [6] speculated that the process
is probably dislocation climb controlled by short circuit
diffusion.
Although there has been extensive work devoted to
the well-known NiAl/Cr(Mo) eutectic system [813]
and the NiAl Hf alloy [1416], it is surprising that
* Corresponding author. Present address: Institute of Metal Re-
search, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang
110016, Peoples Republic of China. Tel.: +86-24-2384-3531; fax:
+86-24-2389-1320.
E-mail address: [email protected] (C.Y. Cui).
0921-5093/02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved.
PII: S0921- 5093( 01) 01452- 6
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 187
relatively scarce investigation has studied the effect of
element Hf on the NiAl/Cr(Mo) alloy. Since Hf is a
very effective solid-solution strengther for NiAl, one
would expect that the NiAl/Cr(Mo,Hf) composite will
exhibit better fracture toughness and elevated tempera-
tures properties. Therefore, the present research was
undertaken to examine the microstructure and mechan-
ical properties of directionally solidied (DS) Ni
33Al 28Cr5.5Mo0.5Hf composite. The effect of
strain rate and temperature on tensile ductility and
fracture behavior will be presented in this paper. From
these results, we may get some insight into the mecha-
nisms responsible for deformation of multiphase NiAl
alloys.
2. Experimental procedure
A vacuum induction melted and drop cast ingot of
Ni 33Al 28Cr5.5Mo0.5Hf (at.%) was directionally
solidied under Ar atmosphere in the Al
2
O
3
SiO
2
ce-
ramic mold by standard Bridgman technique. The
growth rate was :5 mm min
1
and thermal gradient
through the solidliquid interface was :7080 C
cm
1
. The ingot was 16 mm in diameter by 150 mm in
length. Cylindrical compressive specimens, 5 mm di-
ameter by 10 mm length, with the compressive axis
parallel to the growth direction, were electrical dis-
charge machined from the ingot. Compressive tests
were conducted in Gleeble 1500 test machine. The
tensile specimens were ground from the as-grown mate-
rials with gauge dimensions of 2.5216 mm. Tensile
testing was performed at temperatures between 300 and
1323 K in air using AG-250KNE machine at approxi-
mate strain rates of 1.0410
4
s
1
, 1.0410
3
s
1
and 1.0410
2
s
1
.
Transmission electron microscopy (TEM) samples
were prepared by using conventional procedures, in-
cluding cutting discs perpendicular to the growth direc-
tion with thickness of 0.3 mm, mechanical polishing
them to 50 mm thickness, dimpling to 20 mm and nally
ion milling. TEM observations were conducted in a
JEM2010 high resolution electron microscope. Scan-
ning electron microscopy (SEM) analysis was per-
formed in JSM6301F. Secondary electron imaging was
often used to observe the composites microstructure.
3. Results and discussion
The microstructure of longitudinal and transverse
section of DS NiAl 28Cr5.5Mo0.5Hf is shown in
Fig. 1(a) and (b), respectively. Fig. 1(a) shows the
typical microstructure through the transverse section,
which is composed of a lamellar Cr(Mo) phase, NiAl
matrix and a distributed white phase. Energy dispersive
X-ray spectroscopy (EDXS) analysis of this white phase
revealed the composition is: Al 23.48, Ni 48.78, Hf
21.46, Si 3.69, Cr 2.59, in which the Ni:Al:Hf=2:1:1
ratio indicates the white phase is the Heusler phase
(Ni
2
AlHf) with a low level of Si and Cr. It should be
noted that Si is a contamination resulting from the use
of an Al
2
O
3
SiO
2
ceramic mold during the directional
solidication process. The Heusler phase is a ternary
compound, with a cubic L2
1
lattice structure in which
the unit cell comprises eight b.c.c. cells with Al and Hf
atoms occupying two sets of octahedral sites located at
body-center sublattice position [14]. The lattice parame-
ter of Heusler phase is 0.6081 nm, which is close to two
times that of the NiAl. As shown in Fig. 1(b), the
microstructure of the longitudinal direction is charac-
terized by a lamellar reinforcing morphology that is
aligned parallel to the growth direction and embedded
with a NiAl matrix. The selected area diffraction pat-
tern of NiAl and Cr(Mo) established a cube-on-cube
orientation relationship between NiAl matrix and
Cr(Mo) phase, i.e. [111]
NiAl
//[111]
Cr(Mo)
, ()
NiAl
//()
Cr(Mo)
,
which is consistent with previously reported results [17].
Diffraction patterns parallel to the growth direction
also indicate a 111 growth direction for both phases.
Fig. 2(a) is a TEM micrograph, showing a ne array
of precipitates with an average size of 25 nm in the
Fig. 1. The microstructures of longitudinal and transverse section of DS NiAl 28Cr5.5Mo0.5Hf alloy. (a) A typical microstructure through
the transverse section. (b) The microstructure of the longitudinal direction. H denotes Ni
2
AlHf.
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 188
Fig. 2. (a) TEM graph of the ne coherent NiAl precipitates presented in Cr(Mo) phase. (b) TEM graph of ne Cr(Mo) precipitates with regular
arrangement embedded within the NiAl matrix.
Fig. 3. (a) TEM morphology of Heusler phase at the NiAl/Cr(Mo) interface (b) corresponding diffraction pattern of Ni
2
AlHf and NiAl phase.
Cr(Mo) phase. The EDXS results of these ne precipi-
tates are as follows (at.%): Al 41.29, Ni 56.35, Cr 1.93,
Mo 0.43, which indicate that the ne precipitates are
Ni-rich NiAl. The Cr(Mo) phase contains NiAl precipi-
tates indicating some solubility of Ni and Al in the
Cr(Mo) phase at high temperatures. In addition to the
lamellar Cr(Mo) phases embedding within the NiAl
matrix, ne Cr(Mo) precipitates with an average size of
30 nm were also observed in the NiAl matrix (Fig. 2b).
Fig. 3(a) shows the block white phase at the NiAl and
Cr(Mo) interface. TEM results conrmed this phase to
be a Heulser phase as determined from selected area
diffraction patterns (Fig. 3b). Because of the Si contam-
ination, the G-phase precipitate was also found in the
composite. There are many cuboidal precipitate phases
in the vicinity of the Heusler phase, as shown in Fig.
4(a). Selected area diffraction patterns (Fig. 4b) con-
rmed this phase to be the G-phase. The ne G-phase
precipitates, ranging from 5 to 20 nm in size, are also
cuboidal in shape. As reported earlier [18], substantial
Si, available from the Al
2
O
3
SiO
2
shell mold, has dis-
solved in the molten alloy, resulting in the formation of
the ne G-phase precipitate during the solidied
process.
According to NiAl CrMo phase diagram [9] and
NiAl Ni
2
AlHf pseudobinary phase diagram [14], the
solidicated process of NiAl 28Cr5.5Mo0.5Hf com-
posite could be deduced as follows: The NiAl and
Cr(Mo) eutectic reaction occurs at :1718 K rstly,
LNiAl +Cr(Mo) +L
1
(1)
and the remaining liquid (L
1
) becomes enriched in Hf
and Si. Then the intercellular Heusler phase solidies
directly from the liquid as a ternary peritectic reaction
L
1
+NiAl +Cr(Mo) Ni
2
AlHf (2)
Then, G-phase precipitates from the solid state by
diffusion. Therefore, large Heusler particles along with
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 189
Fig. 4. (a) Fine G-phase precipitate in the vicinity of Heusler phase (b) corresponding diffraction pattern of G-phase and NiAl phase. G denotes
Ni
16
Hf
6
Si
7
.
ne G-phase precipitates co-exist near the NiAl/Cr(Mo)
interface. When the content of Hf and Si is large, at
high solidicated rates in some region, the Heusler
phase and G-phase precipitates can grow into large
blocks by the diffusion of Si, Hf and Al.
The effect of strain rate on the tensile ductility (plas-
tic strain to failure) of this multiphase NiAl alloy is
illustrated in Fig. 5, which is a plot of tensile ductility
versus temperature at strain rates of 1.0410
4
,
1.0410
3
and 1.0410
2
s
1
. For clarity, the
BDTT in this analysis is dened as the lowest tempera-
ture for which 5% tensile ductility could be consistently
achieved in the alloy. The stressstrain response indi-
cates that the alloy exhibits a brittle characteristic at
room temperature up to 1173 K at the test strain rate.
Above 1173 K, the BDTT increases signicantly with
an additional increase in the temperature, which shows
that the alloy is ductile at elevated temperature. The
BDTT of the present alloy is higher than 1173 K at the
strain rate of 1.0410
4
s
1
. In addition, the BDTT
of the alloy is found to be sensitive to the strain rate.
The data at high strain rates resemble the data at the
strain rate of 1.0410
4
s
1
but the ductility versus
temperature curves are displaced to higher temperatures
as the strain rate is increased. The BDTT of DS NiAl/
Cr(Mo, Hf) alloy becomes shifted :120 K because of
the two-order-of magnitude increase in the strain rate.
The yield stress at the strain rate of 1.0410
4
s
1
,
1173 K was high (410 MPa) and gradually decreased
with increasing test temperature (Fig. 6). While the
fracture strength at the same strain rate increased
slightly with increasing test temperature up to 1173 K
and then reduced dramatically with increasing test tem-
perature. Recent work [6] by Noebe et al. showed that
the fracture stress and fracture toughness increased
with increasing test temperature, which is contrary to
the present work.
Fig. 7 shows typical true stressstrain graphs near
BDTT. Below the BDTT, signicant work hardening
can be observed after yielding, whilst above BDTT,
Fig. 5. The effect of strain rate and temperature on the tensile
ductility of DS NiAl/Cr(Mo,Hf) alloy.
Fig. 6. Effect of strain rate on the tensile yield strength and fracture
stress of DS NiAl/Cr(Mo,Hf) alloy.
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 190
Fig. 7. Typically true stressstrain graph of DS NiAl/Cr(Mo,Hf)
alloy at near BDTT.
material is associated with this process in the NiAl
phase.
Besides changes in ductility and fracture stress at the
brittle-to-ductile transition temperature, a signicant
change in the fracture morphology is observed just
above BDTT (Fig. 8). The tensile fracture behavior of
the present alloy is brittle with NiAls cleavage and
debonding along the NiAl/Cr(Mo) interface at the
strain rate of 1.0410
4
s
1
and below 1173 K (Fig.
8a). Beyond 1173 K, the fracture behavior is predomi-
nately ductile in nature with extensive necking and a
dimple fracture surface at 1203 K and the same strain
rate (Fig. 8b). The BDTT is improved with the increase
of strain rate. The fracture behavior is also NiAls
cleavage and debonding along the NiAl and Cr(Mo)
interface at 1223 K and the strain rate of 1.0410
3
s
1
(Fig. 8c), which mainly occurs at dendritic arm of
the eutectic alloy. By 1238 K at this strain rate, the
fracture morphology has changed to predominantly
ductile in nature (Fig. 8d). At the strain rate of 1.04
10
2
s
1
, the same results are observed once again.
Therefore, the fracture mode of multiphase NiAl
28Cr5.5Mo0.5Hf changes from debonding along the
NiAl and Cr(Mo) interface and NiAls cleavage below
the BDTT to completely ductile fracture above the
BDTT.
Failure analysis of DS NiAl/Cr(Mo) [12] indicates
that the tensile damage was initiated at the junction of
eutectic grain (bundles) with misaligned lamellar
strain softening is steadily observed after a peak stress
with slight necking before failure, which is not like pure
NiAl tensile graph [5]. The NiAl/Cr(Mo,Hf) alloy is
brittle from room temperature to 1173 K at the test
strain rate, however, the samples tested at 1203 K show
extensive necking and 15% strain before its failure at
the strain rate of 1.0410
4
s
1
. The curves at other
strain rates resemble the curve at the strain rate of
1.0410
4
s
1
. For most materials, the strain soften-
ing during elevated temperature deformation generally
corresponds to the occurrence of dynamic recovery and
recrystallization, it appears the BDT process of this
Fig. 8. The fracture morphology of NiAl/Cr(Mo,Hf) alloy tested at (a) 1173 K and initial strain rate of 1.0410
4
s
1
; (b) 1203 K and initial
strain rate of 1.0410
4
s
1
(microvoid coalescence); (c) 1223 K and initial strain rate of 1.0410
3
s
1
; (d) 1238 K and initial strain rate
of 1.0410
3
s
1
(microvoid coalescence). represents NiAls cleavage. represents debonding along the NiAl/Cr(Mo) interface.
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 191
boundary or at the growth defect of the lamellar struc-
ture. Similar fracture morphology has also been ob-
served in a two-phase titanium aluminide alloy with
lamellar structure [19]. The ductility of multiphase Ni
20Al 30Fe alloy is attributed to: (1) the inhibition of
crack nucleation in the i phase due to enhanced plas-
ticity afforded by the strain transfer from the ductile
second phase [20,21]; and (2) the inhibition of crack
propagation by plastic stretching of the ductile second
phase in the crack wake [22]. However, if the interface
is weak, the stress concentration due to the constrained
deformation of the ductile phases may nucleate cracks
at the interface and thus inhibit operation of such a
mechanism, such as CoNi Al Ti i+k/k% multiphase
alloys [23]. As to the present alloy, there exist a lot of
element Hf in the fracture morphology from the energy
dispersive X-ray spectroscopy (EDXS). From the frac-
ture morphology presented in this study, we suggest
that the low ductility of the present alloy should be
attributed to the Heusler phase existed at the NiAl/
Cr(Mo) interface, which reduce the bonding strength of
the interface, inhibit operation of strain transfer and
facilitate crack propagation along the NiAl/Cr(Mo)
interface.
It is well established that the strength of NiAl alloy is
very sensitive to deviation from stoichiometry and
much work has demonstrated that deviations from
stoichiometry can signicantly improve the yield
strength [24] and BDTT [25]. The composition of NiAl
phase is as follows by EDXS: Al 43.10, Ni 47.94, Cr
7.62, Hf 0.43, Mo 0.91, which shows that the alloy is
off-stoichiometry. Considering the strengthening effects
due to such factors as deviation from stoichiometry,
Cr(Mo) phase and Heusler phase precipitates in the
NiAl matrix, we suspect that the consistent phase below
BDTT (NiAl phase, Cr(Mo) lamellar phase and inter-
cellular Heulser phase), each inherently exhibits higher
strength than bonding strength NiAl and Cr(Mo) inter-
face. In addition, since single NiAl phase holds its
strength very well below BDTT and its BDTT signi-
cantly increases with the increasing of strain rate and
the addition of other element, such as Ti. As a result,
when the alloy is tested in tension at low temperature
or higher strain rate, debonding failure along the NiAl
and Cr(Mo) interface might occur before the consistent
phases reached their fracture strength. The same result
is concluded from other multiphase NiAl alloy like
Ni 25Al 15Cr alloy [26].
The effect of strain rate on the BDTT for NiAl
would be expected to have one of the two conse-
quences, depending on the mechanism responsible for
the transition from brittle to ductile behavior with
increasing temperature. If the change is due to the
operation of additional slip systems at some critical
temperature, for the operation of non-a001 slip vec-
tors can afford additional slip system and be consistent
with the Von Mises criterion, the additional energy for
diffusion and thermally activated process is not needed
for dislocation climb and grain slide, then the BDTT
should be independent of strain rate. If incompatibility
at the grain boundaries is relieved through the combi-
nation of glide and climb of 100 type dislocations as
previously proposed, the BDTT would be sensitive to
deformation rate. The results of the present alloy
demonstrate that changes in strain rate do inuence the
tensile properties of polycrystalline NiAl in a manner
consistent with this latter explanation, which suggested
that a rate-dependent deformation mechanism is re-
sponsible for the ductility observed at elevated
temperature.
At elevated temperature, tensile ductility is not a
problem, since the alloy undergoes a dramatic brittle-
to-ductile transition over a temperature range of 1178
1303 K, depending on the strain rate. Strain rate
supports the notion of diffusive mechanisms providing
the additional deformation mechanisms at the BDTT in
NiAl, as shown in Fig. 9(a). At and above 1123 K, the
compressive yield stress of NiAl/Cr(Mo,Hf) is seen to
be a function of strain rate, whereas at 1123 K, negligi-
ble strain rate dependence is observed. The results
imply that a thermally activated process is contributing
to the deformation of NiAl at temperatures above
:1123 K. Since yield stress was dependent on tempera-
ture, it is appearing to represent the data in an Arrhe-
nius form, as shown in Fig. 9(b), i.e.
|
ya
=A e
Q
1
/RT
(3)
where A is a constant, R is the universal gas constant,
Q is an activation energy and T is the absolute temper-
ature. In Fig. 9(b), a change in slope implies a change
in mechanism. The BDTT corresponds to have signi-
cant ductility in this material. From analysis of these
data, Q
1
was found to equal :486 kJ mol
1
. A map
dening the conditions for which the present alloy is
ductile or brittle was obtained by plotting strain rate
versus BDTT (Fig. 9c). Above and to the left of the
line, the present alloy exhibits low ductility and fails in
a brittle fashion. Since the BDTT for NiAl increases
with increasing strain rate, the additional deformation
mechanisms responsible for the BDT must also be rate
dependent. Consequently, the data presented in Fig.
9(c) can be tted to an Arrhenius equation of the form,
m; =B e
Q
2
/RT
(4)
where is the strain rate, Q
2
is the apparent activation
energy for the deformation over the range of the
BDTT, T is the brittle-to-ductile transition temperature
and B is a pre-exponential constant. From analysis of
these data, Q
2
was found to equal :463 kJ mol
1
.
The Q
2
was almost the same as Q
1
calculated from the
stress versus temperature behavior.
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 192
Fig. 9. (a) Effect of strain rate on compressive ow stress of NiAl/
Cr(Mo,Hf) as a function of temperature. (b) Arrhenius representation
of compressive yield stress as a function of temperature. (c) Strain
rate versus temperature map indicating the strain rate-temperature
conditions for which DS NiAl/Cr(Mo,Hf) is brittle and ductile.
absolute melting temperature of the material). The acti-
vation energy calculated for lattice diffusion in NiAl is
220300 kJ mol
1
[6]. The apparent activation energy
calculated from this study is much higher than the
value of lattice diffusion in NiAl. There are many
precipitating phases (Cr(Mo), Heusler phase and G-
phase precipitate) in the NiAl matrix, which impede the
dislocation movement. In other alloy containing the
element of Hf, Ti in AFN-20 [27], the BDTT is :1423
K, while the BDTT of NiAl Ta eutectic alloy investi-
gated by Zeumer [28] is :1473 K at a strain rate of
10
4
s
1
, which is higher than the present alloy.
Therefore, the element Hf can affect the onset of climb
by forming solute atmospheres around mobile disloca-
tions and effectively pinning them, additional energy
(higher temperature) would be necessary to overcome
this pinning effect. A recent examination [29] of dislo-
cation pinning by solute (Hf, Zr, C and B) in NiAl has
concluded that low temperature improvements in yield
strength are the results of Cottrell atmosphere pinning
mobile dislocations. Also, some evidences supporting
this argument have been presented in deformation stud-
ies of NiAl:Zr [30]. Therefore, the higher BDTT can be
gained with the addition of solid strengthening element,
such as Hf, Zr, Ti and Ta.
The apparent activation energy of the as-grown com-
posite for the BDTT is :463 kJ mol
1
, which is the
same as the activation energy for deformation at the
temperature range of 12001400 K (Q=456.6 kJ
mol
1
) investigated by Johnson et al. [10]. In disagree-
ment with the present result, previous investigation by
Noebe et al. [6] showed that the mechanism for the
brittle-to-ductile transition in stoichiometric NiAl is
associated with dislocation climb controlled by short
circuit diffusion with the activation energy of 118 kJ
mol
1
, while the multiphase Ni 25Al 25Cr investi-
gated by Chen et al. [26] has the activation energy of
253 kJ mol
1
, suggesting the mechanism be associated
with lattice diffusion in NiAl phase, which is not suit-
able for the present composite with high BDTT. Much
work needs to be done before a profound understand-
ing of the mechanism for the BDT of multiphase NiAl
alloys can be suggested.
4. Conclusions
Effects of temperature and strain rate on the tensile
behavior of multiphase NiAl alloys are evaluated.
Based on the results of present work, the following
conclusions have been made.
The microstructure of DS NiAl/Cr(Mo,Hf) alloy
consists of NiAl matrix, lamellar Cr(Mo) phase and
intercellular Heusler phase. There exist Cr(Mo) precipi-
tates, Heusler phase and G-phase precipitates in the
NiAl matrix.
The activation energy for deformation or creep of
metallic generally corresponds to the activation energy
for self-diffusion at temperature \0.5 T
m
(T
m
is the
C.Y. Cui et al. / Materials Science and Engineering A325 (2002) 186193 193
The BDTT of the present alloy is determined to be
:1178 K at a strain rate of 1.0410
4
s
1
and
dependent on the strain rate, with a two-order-of -mag-
nitude increase in strain rate resulting in :120 K
increase in BDTT.
At temperature above BDTT, the fracture morphol-
ogy changes from cleavage of NiAl and debonding
along the NiAl/Cr(Mo) interface to microvoid coales-
cence in nature. The precipitating phase (Cr(Mo),
Heusler phase and G-phase precipitates) in the NiAl
matrix is responsible for the high BDTT.
Acknowledgements
This work was supported by the National Natural
Science Foundation of China (No. 59895152) and Na-
tional Advanced Materials committees of China (863-
715-005-0030) to whom we are very grateful.
References
[1] R. Darolia, J.J. Lewandowski, C.T. Liu, P.L. Martin, D.B.
Miracle, M.V. Nathal (Eds.), Structural Intermetallics. The Min-
erals, Metals and Materials Society, Warrendale, PN, 1993.
[2] A. Ball, R.E. Smallman, Acta Metall. 14 (1966) 1349.
[3] A. Ball, R.E. Smallman, Acta Metall. 14 (1966) 1517.
[4] F. Ebrahimi, S. Shrivastava, Acta Mater. 46 (1998) 1493.
[5] F. Ebrahimi, G.T. Hoyle, Acta Mater. 45 (1997) 4193.
[6] R.D. Noebe, C.L. Cullers, R.R. Bowman, J. Mater. Res. 7
(1992) 605.
[7] D. Miracle, Acta Metall. Mater. 39 (1991) 1457.
[8] H.E. Cline, J.L. Walter, E. Lifshin, R.R. Russell, Metall. Trans.
A 2 (1971) 189.
[9] H.E. Cline, J.L. Walter, Metall. Trans. A 1 (1970) 2907.
[10] D.R. Johnson, X.F. Chen, B.F. Oliver, R.D. Noebe, J.D. Whit-
tenberger, Intermet 3 (1995) 99.
[11] X.F. Chen, D.R. Johnson, R.D. Noebe, B.F. Oliver, J. Mater.
Res. 10 (1995) 1159.
[12] J.M. Yang, S.M. Jeng, K. Bain, R.A. Amato, Acta Mater. 45
(1997) 295.
[13] D.T. Jiang, J.T. Guo, Mater. Sci. Eng. A 255 (1998) 154.
[14] M. Takeyama, C.T. Liu, J. Mater. Res. 5 (1990) 1189.
[15] J.D. Whittenberger, I.E. Locci, R. Darolia, R.R. Bowman,
Mater. Sci. Eng. A 268 (1999) 165.
[16] R. Darolia, D. Lahrman, R. Field, Scr. Metall. Mater. 269
(1992) 1007.
[17] H.E. Cline, J.L. Walter, E.F. Koch, L.M. Osika, Acta Metall.
Mater. 19 (1971) 405.
[18] I.E. Locci, R.M. Dickerson, A. Grag, R.D. Noebe, J.D. Whit-
tenberger, M.V. Nathal, R. Darolia, J. Mater. Res. 11 (12)
(1996) 3204.
[19] K.S. Chan, Y.W. Kim, in: Y.W. Kim, R.R. Boyor (Eds.),
Microstructure Property Relationships in Titanium Aluminides
and Alloys, TMS, Warrendale, PA, 1991, p. 179.
[20] S.C. Huang, R.D. Field, D.D. Krueger, Metall. Mater. Trans. A
21 (1990) 959.
[21] R.D. Noebe, A. Misra, R. Gibala, ISIJ 31 (10) (1991) 1172.
[22] S. Guha, I. Baker, J. Mater. Sci. 31 (1996) 4055.
[23] Y. Mishima, E.H. Lee, C.T. Liu, Mater. Trans. JIM 36 (8)
(1995) 1030.
[24] R.T. Pascoe, C.W. Newey, J. Met. Sci. 2 (1968) 138.
[25] R. Bowman, R. Noebe, R. Darolia, NASA Conf. Pub. 10039
(1989) 47(1) HITEMP review.
[26] R.S. Chen, J.T. Guo, W.L. Zhou, J.Y. Zhou, Intermet 8 (2000)
663.
[27] R. Darolia, W.S. Walston, in: M.V. Nathal, R. Darolia, C.T.
Liu, P.L. Martin, D.B. Miracle, R. Wagner, M. Yamaguchi
(Eds.), Structural Intermetallics, The Minerals, Metals and Ma-
terials Society, Warrendale, PA, 1997, p. 585.
[28] B. Zeumer, G. Sauthoff, Intermet 6 (1998) 451.
[29] S. Lee, J.C.M. Li, C.T. Liu, Mater. Sci. Eng. A 239/240 (1997)
808.
[30] R.D. Noebe, R.R. Bowman, C.L. Cullers, S.V. Ray, Third
Annual HITEMP Review, 1990, NASACP-10051, (1990) 2021.