Available online at www.sciencedirect.

com

Acta Materialia 58 (2010) 64116420

www.elsevier.com/locate/actamat

Role of severe plastic deformation on the cyclic reversibility

of a Ti50.3Ni33.7Pd16 high temperature shape memory alloy
B. Kockar a,c, K.C. Atli a, J. Ma a, M. Haouaoui a, I. Karaman a,,
M. Nagasako b, R. Kainuma b
a

Department of Mechanical Engineering, Texas A&M University, MS 3123, College Station, TX 77843, USA
b
Department of Materials Science, Tohoku University, Sendai 980-8579, Japan
c
Department of Mechanical Engineering, Hacettepe University, Beytepe, 06800 Ankara, Turkey
Received 7 June 2010; received in revised form 3 August 2010; accepted 3 August 2010

Abstract
The present work focuses on the eect of microstructural renement on the thermo-mechanical cyclic stability of a Ti50.3Ni33.7Pd16
high temperature shape memory alloy (HTSMA) which was severe plastically deformed using equal channel angular extrusion (ECAE).
The grain/subgrain size of the high temperature austenite phase was rened down to about 100 nm, the lowest reported to date in HTSMAs. The increase in strength dierential between the onset of transformation and the macroscopic plastic yielding after ECAE led to a
notable enhancement in the cyclic stability during isobaric coolingheating experiments. The reduction in irrecoverable strain levels was
attributed to the increase in critical stress for dislocation slip due to the microstructural renement during the ECAE process.
2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Keywords: TiNiPd; High temperature shape memory alloys; Severe plastic deformation; Equal channel angular extrusion; Equal channel angular pressing

1. Introduction
NiTi shape memory alloys (SMAs) have revolutionized
the development and use of active materials in the last
40 years by providing large reversible shape changes, high
actuation forces, and large elastic strains as a result of thermoelastic martensitic transformations. The development of
SMA actuators has recently been a priority in space, automotive and power generation applications, where high
operating temperatures are needed. However, the low operation temperatures (<100 C) and relatively large transformation hysteresis of binary NiTi limit their utility in high
temperature applications [1]. Luckily, the transformation
temperatures of NiTi can be controlled by the addition of
alloying elements. For instance, Pd addition at more than

Corresponding author.

E-mail address: ikaraman@tamu.edu (I. Karaman).

10 at.% increases the transformation temperatures to above

100 C [2,3].
NiTi-based ternary alloys with the addition of one of
Pd, Pt, Zr, Au, or Hf have been the most studied high temperature shape memory alloys (HTSMAs) to date [25].
The addition of Zr and Hf could be more favorable as a
third alloying element due to their low cost, however, NiTi
alloys with Zr and Hf exhibit unstable shape memory properties and their thermal and stress hystereses are large [4,5].
The addition of Pt to a NiTi binary alloy would massively
increase the cost. Therefore, Pd could be a better choice as
a third alloying element considering the increase in transformation temperatures and the decrease in thermal hysteresis [2,6].
Dislocation plasticity that may accompany martensitic
transformation is an important anticipated problem in
NiTiPd HTSMAs, aecting the stability of the shape
memory eect and superelasticity. A low critical stress for
dislocation slip at high temperatures can easily cause the

1359-6454/$36.00 2010 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.actamat.2010.08.003

6412

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

accommodation of transformation shear and volume

change by dislocation slip, and cause dissipation of stored
elastic energy during transformation. Eventually, this leads
to an apparent permanent deformation, dimensional instability, and a lack of superelasticity [7,8]. Well-developed
dislocation substructures, desired crystallographic texture,
rened grain sizes, and coherent precipitates, all of which
improve the critical stress for slip (CSS), have been
reported as necessary conditions for cyclic and dimensional
stability and for superelasticity in HTSMAs [9,10]. Four
strategies have been proposed to achieve an increase in
CSS: (1) addition of a quaternary element, (2) age hardening of the parent phase, (3) thermo-mechanical treatment
and (4) severe plastic deformation (SPD) [10].
A limited number of studies have reported on the
mechanical and shape memory properties of NiTiPd alloys.
Cai et al. [7] showed that the shape memory behavior of
TiNiPd alloys was fairly good when deformed at room
temperature, however, became relatively poor with increasing deformation temperature in martensite due to a
decrease in CSS. Lindquist and Wayman reported 6%
unconstrained shape recovery [6]. Khachin et al. [11]
showed 4% complete strain recovery, which was introduced
by applying 200 MPa torsional stress in Ni13Ti50Pd37.
Otsuka and co-workers studied the shape memory eect
in Ti50Pd50 [8]. They reported poor shape memory behavior, which could again be attributed to a low CSS possibly
leading to a high density of slip upon deformation, in addition to the twinning in martensite [8]. NiTi alloys with 40
50 at.% Pd show only 0.5% shape recovery when loaded in
tension [1].
Several attempts at enhancing the shape memory
response of NiTiPd alloys have been reported. Yang and
Mikkola [12] examined the eect of boron addition on
the shape memory characteristics of a Ni22.3Ti50.7Pd27 alloy
and found 90% shape recovery for 23% applied strain
under compression. Boron had no real eect on the shape
memory characteristics. However, they determined an
increase in ductility at room temperature and attributed
this to grain renement due to boron addition [12].
Another method to improve the shape memory characteristics of the NiTiPd alloys is to use slightly o-stoichiometric compositions and generate homogeneously distributed
Ti2Ni precipitates in a Ni19.4Ti50.6Pd30 alloy [13]. 90%
shape recovery was observed when a Ni19.4Ti50.6Pd30 alloy
was deformed to a total strain of 6% at 473 K. This was
about 10% higher than that of Ti50Ni30Pd20, in which precipitation was not observed. Such an increase was attributed to the hardening eect of the homogeneously
distributed Ti2Ni precipitates [13]. Moreover, certain
thermo-mechanical treatments have been applied to modify
the microstructure and strengthen the material via grain
renement and dislocation hardening. Golberg and coworkers reported some improvement in the shape memory
properties of Ti50Ni30Pd20 after cold rolling and subsequent heat treatment [14,15]. They claimed superelasticity
for the rst time in Ti50Ni30Pd20 after annealing the alloy

at 673 K and testing at 535 K [14,15]. Cai et al. [7] studied

the eect of thermal cycling on the shape memory properties of Ni19.4Ti50.6Pd30 and found that Ms temperature,
total transformation strain, and irrecoverable strain
increased with the number of cycles. The change in these
parameters occurred quickly in the rst 40 cycles and then
tended to stabilize [7].
As an alternative to cold working, precipitation hardening and thermo-mechanical training, microstructural renement using severe plastic deformation (SPD) should also
bring about good cyclic reversibility and shape recovery in
NiTiPd HTSMAs. SPD via high pressure torsion (HPT)
leads to the formation of nanosized grains and amorphization in bulk binary NiTi samples, however, HPT samples
are too small to investigate the shape memory properties
of the alloys [1618]. On the other hand, SPD via equal channel angular extrusion (ECAE) has several advantages over
HPT, such as producing larger samples, introducing uniform deformation, and allowing some control over grain
morphology and crystallographic texture. Our previous
studies on ECAE of Ni-rich NiTi, equiatomic NiTi and TiNiHf HTSMAs have shown that SPD via ECAE enhances the
cyclic stability and shape recovery behavior under relatively
high stress levels by reducing the grain size and inducing ne
dislocation substructures [10,1922].
Most of the aforementioned studies on TiNiPd alloys
demonstrated shape recovery of these alloys under stressfree conditions, and presented the isothermal mechanical
properties and microstructure. Most applications, however, require shape recovery under applied stress and/or
thermal cyclic stability under constant loads. Only one
research group has recently reported the shape memory
response of Ni19.5Ti50.5Pd30 HTSMAs and corresponding
work outputs during few thermal cycles under various
loads, who observed irrecoverable strains at relatively high
stress levels [23,24].
Thus, this work presents the rst report on the SPD of a
Ti50.3Ni33.7Pd16 alloy produced by ECAE. We introduce
the inuence of ECAE on the evolution of microstructure,
and cyclic reversibility and shape recovery during thermal
cycling under stress. The material was ECAE processed
at 425 C following route E for four passes. The results
show that the cyclic stability was dramatically enhanced
after the ECAE process.
2. Experimental procedure
The Ti50.3Ni33.7Pd16 alloy was prepared by vacuum
induction melting. A cylindrical billet was then hot rolled
at 925 C down to 1.2 cm diameter. 10 or 12.5 cm long bars
were canned in 304 stainless steel (SS) before ECAE. Both
Ti (commercial purity grade 2) and 316 SS were explored as
canning materials prior to using 304 SS. Ti was too soft at
the processing temperature and caused the enclosed SMA
billets to undergo severe shear localization during multipass ECAE. The high strain hardening of 316 SS at the
extrusion temperature substantially increased the press

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

load after the second or third ECAE pass. 304 SS experienced lower strain hardening and thus proved to be the
right choice, given the limited temperature window for
extrusion imposed by these hard to work SMAs and the
load limit of the ECAE die at high extrusion temperatures.
The canned billets were kept for 30 min in the ECAE die
before the rst pass and for 15 min prior to successive
passes to allow sucient heating and to minimize the
recovery/recrystallization as much as possible. Before the
rst pass and between the successive passes the die and billet were lubricated to reduce the friction between them and
between the dierent moving parts of the die. The whole
set-up was enclosed in a furnace which was preheated to
425 C. The extrusions were conducted in a nitrogen atmosphere to minimize potential oxidation. ECAE was conducted following route E, having a billet rotation
sequence of +180, 90, 180 at 425 C. Route E was
selected because it results in the largest volume of uniformly processed material (i.e. the largest product yield)
and the greatest number of equiaxed grains [25]. 425 C
was the lowest temperature at which samples without
micro/macro shear localization could be produced. Four
ECAE passes were performed for each billet, since at least
four or a higher number of passes is required to obtain a
signicant volume fraction of high angle grain boundaries
in bcc structures during ECAE [26,27]. The billet was water
quenched after each ECAE pass to minimize recovery during slow cooling. Fig. 1 shows one of the processed billets
that had a 10 cm processed length.
As-received and ECAE processed specimens were thermally cycled ve times between 20 C and 200 C using a
PerkinElmer Pyris I dierential scanning calorimeter.
The heatingcooling rate during these experiments was
set to 10 C min1.
Transmission electron microscopy (TEM) studies were
conducted using a JEOL JEM-2000 EX-II microscope
operated at an accelerating voltage of 200 kV. The TEM
samples were mechanically ground down to a thickness
of 100 lm and twin jet electro-polished with a 8 vol.% perchloric acid, 72 vol.% acetic acid, 12 vol.% ethanol and
8 vol.% ethylene glycol solution at room temperature.
Bright eld images and diraction patterns of the asreceived and ECAE processed samples were recorded at
room temperature to investigate the martensite twin struc-

6413

ture. In order to examine the austenite phase, the samples

were heated to above the austenite nish (Af) temperature
using an in situ TEM heating stage.
The isothermal monotonic loading experiments under
compression and thermo-mechanical characterization
experiments under tension were conducted using an MTS
810 servo-hydraulic test frame. Cooling and heating of
the samples was achieved by conduction through the grips,
which were heated using heating bands and cooled using
liquid nitrogen owing through copper tubing wrapped
around the grips. The heatingcooling rate was
10 C min1. A high-temperature extensometer with a
gauge length of 12.54 mm was used to record the axial
strains by attaching its ceramic extension rods directly to
the gage section of the tensile samples. Compression specimens with the dimensions of 4  4  8 mm and small dogbone-shaped tension samples with a gauge section of
1.5  3  8 mm were cut using wire electro-discharge
machining (EDM) from the uniformly deformed region
of the billets. For the compression experiments a miniature
extensometer with a 3 mm gage length was used to measure
the strain directly on the sample. The temperature was
measured using a thermocouple attached to the samples.
Three types of experiments were conducted on the asreceived and ECAE processed samples:
1. Isothermal monotonic loading experiments under compression in austenite to determine the eect of ECAE
on the strength levels. The test temperature was 20 C
above the martensite start temperature (Ms) of each
MPa
sample measured under a 50 MPa load (M r50
),
s
which was selected to ensure thermodynamically equivalent conditions for the two dierent microstructures.
The strain rate was 5  104 s1.
2. Isobaric coolingheating experiments under increasing
tensile stress levels with 50 MPa increments to determine
the shape memory characteristics, such as transformation temperatures, recovered transformation and irrecoverable strain levels as a function of stress, and to
determine the threshold stress level for the onset of irrecoverable strain.
3. Isobaric coolingheating experiments up to 10 cycles
under 150 MPa load to evaluate the cyclic stability of
the as-received and ECAE processed samples.

Fig. 1. The digital image of the ECAE processed Ti50.3Ni33.7Pd16 high temperature shape memory alloy billet. The billet was 10 cm long and was ECAE
processed for up to four passes at 425 C using route E.

6414

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

3. Experimental results
3.1. Microstructural evolution
The grain size of the as-received sample was impossible
to determine using optical microscopy at room temperature
since the microstructure consisted of martensite variants
and twins, as shown in Fig. 2. The martensitic structure
did not allow the detection of the prior austenite grain
boundaries. Additionally, revealing the austenite grain
boundaries was impossible in the as-received sample using
in situ TEM after heating the sample above the Af due to
the large grain size. Figs. 2 and 3 show the bright eld
TEM images and corresponding selected area diraction
(SAD) patterns of the as-received and ECAE processed
samples at room temperature in the martensitic state. Both
samples had a B19 orthorhombic structure, as expected for

500nm

(a)

the martensite in these ternary SMAs [2,6]. The ECAE processed sample had a notably smaller martensite twin thicknesses than the as-received sample. Such renement after
ECAE was also observed in binary NiTi [19]. Fig. 4 shows
the austenite structure in the ECAE processed samples at
200 C with an average grain size on the order of 100 nm.
3.2. Dierential scanning calorimetry
Transformation temperatures under stress-free conditions for both cases were determined using dierential
scanning calorimetry (DSC) measurements shown in
Fig. 5. The transformation temperatures of the rst and
fth cycles and the thermal hysteresis of the as-received
and ECAE processed samples are summarized in Table 1.
The Ms temperature of the as-received sample was
93.2 C and that of the ECAE processed sample was about
20 C lower. Thermal hysteresis was around 20 C in both

Incident beam//11 0 B19

(b)

Fig. 2. Room temperature (a) bright eld TEM image and (b) corresponding selected area diraction (SAD) pattern of the as-received Ti50.3Ni33.7Pd16
sample. The martensitic structure was internally twinned B19 orthorhombic with {1 1 1} type 1 twins.

500nm

(a)

Incident beam//101 B19

(b)

Fig. 3. Room temperature (a) bright eld TEM image and (b) corresponding SAD pattern of the Ti50.3Ni33.7Pd16 sample ECAE processed at 425 C for
four passes following route E. The twin thicknesses are much smaller than those in the as-received sample.

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

6415

200

011

200nm

(a)

Incident beam//011B2

(b)

Fig. 4. Bright eld TEM image of the B2 austenite structure at 200 C taken from the Ti50.3Ni33.7Pd16 sample ECAE processed at 425 C to four passes
using route E showing the rened grain size of the order of 100 nm.

processed sample in the fth cycle was approximately

4 C lower than that in the rst cycle.
3.3. Isothermal monotonic uniaxial response under
compression

Fig. 5. DSC responses of the (a) as-received and (b) ECAE processed
Ti50.3Ni33.7Pd16 alloy and the transformation temperatures in the rst
cycle are shown in the graphs.

materials and was much lower than that for binary NiTi
[19]. There was almost no change in the transformation
temperatures of the as-received sample with number of
cycles. Similar observations were made in the ECAE processed case, with the exception of the rst heating cycle.
The rst heating cycle showed martensite stabilization
due to the heavily deformed structure, which caused additional resistance to reverse transformation upon rst heating [6]. The austenite start (As) temperature of the ECAE

Fig. 6 displays the monotonic compression response of

the as-received and ECAE processed Ti50.3Ni33.7Pd16 samples. The experiments were conducted at 20 C above the
Ms temperature of each sample, which was measured under
Mpa
50 MPa load (M r50
). The samples were rst heated
s
above Af to ensure that they were fully austenitic, and then
cooled down to the test temperature. This temperature was
selected to clearly distinguish the regions of stress-induced
martensitic transformation and macroscopic plastic deformation of martensite in the stressstrain response, in addition to the samples with dierent microstructures to be
under thermodynamically similar conditions. The compression test results, rather than tension, are presented here
because both materials failed under tension before they
reached the second plateau in the stressstrain response,
which indicates macroscopic plastic deformation of martensite. The early failure in tension was likely to be due
to the relatively high fraction of carbides in the present ternary alloy.
Fig. 6 clearly shows that the macroscopic yield strength
of martensite, rM
y , increased after ECAE, even though the
critical stress required to induce martensite, rSIM, did not
notably change. This dissimilar trend gave rise to a clear
increase in the strength dierential, Dr, between these
two stress levels upon ECAE. Such an increase should be
a consequence of an increase in the CSS due to work hardening and grain renement. Note that rM
y is not discernible
in Fig. 6 for the case of the ECAE processed sample, even
though the sample was still intact at the high stress levels
during the experiment. This was because of the load limit
of the test frame, which did not allow the application of
higher stresses for the present compression sample geometry. It is likely that rM
y for the ECAE processed sample is

6416

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

Table 1
Martensitic transformation temperatures and thermal hysteresis of the as-received and ECAE processed Ti50.3Ni33.7Pd16 high temperature shape memory
alloy samples at the rst and fth thermal cycles, extracted from the DSC data. A1st and M1st are the transformation peak temperatures during the rst
heating and cooling cycles, respectively.
Material

As1st (C)

Ms1st (C)

As5th (C)

Ms5th (C)

As5thAs1st (C)

Ms5thMs1st (C)

A1stM1st (C)

As-received
ECAE 4E at 425 C

99
61.5

93.2
72.6

99
57.6

92.5
71.4

0
3.9

0.7
1.2

19.3
21.6

1500
1000

= yM SIM

5%
250 MPa
200 MPa
150 MPa
100 MPa

500

50 MPa

2%

40

Strain (%)

Fig. 6. Monotonic compression response of the as-received and ECAE

Mpa
processed Ti50.3Ni33.7Pd16 alloys under compression at (M r50
+
s
20 C). This temperature was selected so that the samples were subjected
to similar thermodynamic conditions.

signicantly higher than 1500 MPa. Regardless of this, the

resistance to plasticity is obviously higher in the ECAE
processed sample, and this should point to an enhanced
transformation reversibility and cyclic stability.
3.4. Isobaric coolingheating experiments
The dimensional and cyclic stability and recovered transformation strain levels of Ti50.3Ni33.7Pd16 are important factors for practical applications when the alloy is acting against
a force. Therefore, isobaric coolingheating experiments
were conducted, during which the test specimen was held
under constant stress while it underwent continuous cooling
and heating at a rate of 10 C min1. After each additional
cooling and heating cycle the stress was increased by
50 MPa increments while in the austenitic state. Fig. 7 shows
the strain vs. temperature response of both the as-received
and ECAE processed samples. Fig. 8 schematically illustrates how the recovered transformation and irrecoverable
strain (eirr) values and transformation temperatures were
determined from the isobaric coolingheating experiments.
The recovered transformation strain was measured at temperature (Af + Ms)/2 between the extrapolated baseline
strain levels of the austenite and martensite phases. Transformation hysteresis was determined at the half of the recovered transformation strain level, as shown in Fig. 8. A
summary of the recovered transformation and irrecoverable
strain levels, and thermal hysteresis is presented in Fig. 9a
and b, respectively. The maximum recovered transformation
strain level reached 3.3% at a stress level of 250 MPa in the
as-received sample, however, this was accompanied by an

80
120
Temperature (C)

160

200

(a)
Ti50.3Ni33.7Pd16
ECAE 4E at 425C

350 MPa

5%

300 MPa

Strain (%)

Ti50.3Ni33.7Pd16
As-Received

300 MPa

Ti50.3Ni33.7Pd16
As-Received
ECAE 4E at 425C

Strain (%)

Stress (MPa)

2000

250 MPa
200 MPa
150 MPa
100 MPa
50 MPa

40

80
120
Temperature (C)

160

200

(b)

Fig. 7. Strain vs. temperature responses of the Ti50.3Ni33.7Pd16 alloy under

various constant stress levels during isobaric coolingheating experiments:
(a) as-received and (b) ECAE 4E at 425 C samples.

irrecoverable strain of 0.57%. In fact, irrecoverable strain

began to appear in the as-received sample even under
100 MPa stress. The eirr levels of the ECAE processed sample
shown in Fig. 9a demonstrate dramatic improvements over
the as-received case. No eirr was observed at 100 MPa and
none appeared until 250 MPa, while the recovered transformation strain was lower by 0.4% at 250 MPa compared with
that of the as-received sample.
Another intriguing observation from the isobaric coolingheating experiments was the lower thermal hysteresis
values in the ECAE processed samples above 200 MPa
load. In addition, the hysteresis was quite stable and almost
independent of the applied stress in the processed samples.
On the other hand, the thermal hysteresis of the as-received
sample increased signicantly with stress, especially above
50 MPa. Thermal hysteresis is an indication of how much
energy dissipation occurs during forward and reverse
transformation. Apparently, ECAE signicantly stabilizes
transformation, such that energy dissipation and, thus,

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

6417

350
300

Recovered
Transformation Strain

Stress (MPa)

Strain (%)

Mf

Thermal
Hysteresis

Ms

250
200
100
0
60

100

150

90 100 110 120

Temperature (C)

130

140

150

200
250
Stress (MPa)

300

Ti50.3Ni33.7Pd16
As-Received

150MPa

Strain (%)

Strain (%)

Irrecoverable Strain
As-Received
ECAE 4E at 425C

50

80

Fig. 10. Eect of temperature on the critical stress for the martensitic
transformation in the ECAE processed and as-received Ti50.3Ni33.7Pd16
alloy.

1%

70

Ti50.3Ni33.7Pd16

Recovered Transformation Strain

As-Received
ECAE 4E at 425C

Ti50.3Ni33.7Pd16
ECAE 4E at 425C
As-Received

50

Irrecoverable
Strain

Fig. 8. Schematic illustration of the recovered transformation and

irrecoverable strain, Ms and Mf temperatures, and thermal hysteresis in
a typical isobaric coolingheating experiment.

-1

7.3 MPa C

150

Temperature (C)

-1

8.4 MPa C

350
0

(a)

40

80
120
Temperature (C)

160

200

(a)

Ti50.3Ni33.7Pd16
As-Received
ECAE 4E at 425C

30

Ti50.3Ni33.7Pd16
ECAE 4E 425C

1%

25

Strain (%)

Thermal Hysteresis (C)

35

20

150MPa

15
10

50

100

150
200
Stress (MPa)

250

300

350

40

(b)

Fig. 9. (a) Recovered transformation and irrecoverable strain levels and

(b) thermal hysteresis of the as-received and ECAE processed
Ti50.3Ni33.7Pd16 alloy as a function of constant tensile stress levels
determined from the isobaric coolingheating experiments.

thermal hysteresis, did not change with increasing stress.

The only exception occurred above 350 MPa, at which
the thermal hysteresis of the ECAE processed sample notably increased.
To reveal the eect of ECAE on the onset of martensitic
transformation, a critical stress vs. temperature diagram
was constructed, as shown in Fig. 10, using the results of
the isobaric coolingheating experiments in Fig. 7. rSIM
strongly depends on temperature, with a positive slope
according to the modied ClaussiusClapeyron (CC) equation. The CC relation is dened as [28]:

80
120
Temperature (C)

160

200

(b)

Fig. 11. Cyclic strain vs. temperature response of Ti50.3Ni33.7Pd16 under

150 MPa load for 10 cycles: (a) as-received and (b) ECAE 4E at 425 C
samples.

dr=dT DH =T 0 etr

where DH is the transformation enthalpy, T0 is the chemical equilibrium temperature for the transformation, and
etr is the transformation strain. dr/dT values were determined to be 8.4 and 7.3 MPa C1 for the ECAE processed
and as-received samples, respectively.
3.5. Thermal cycling experiments under constant stress
In thermal cycling experiments, the samples were held at
150 MPa for 10 thermal cycles with a heatingcooling rate

6418

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

6

Ti50.3Ni33.7Pd16

Strain (%)

Recovered Transformation Strain

As-Received
ECAE 4E at 425C

Cumulative Irrecoverable Strain

Strain - As-Received
ECAE 4E at 425C

3
2
1
0

5
6
7
Number of Cycles

10

(a)
Ms Temperature (C)

130

Martensite Start Temperature

As-Received
ECAE 4E at 425C

120

Ti50.3Ni33.7Pd16

110
100
90
80
70

4
5
6
7
Number of Cycles

10

(b)

Fig. 12. (a) Recovered transformation and cumulative irrecoverable strain

and (b) martensite start temperature (Ms) evolution of the as-received and
ECAE processed Ti50.3Ni33.7Pd16 alloy as a function of number of cycles
under 150 MPa load.

structural renement on the thermo-mechanical cyclic stability of a ternary TiNiPd alloy. The main observations can
be summarized as follows:
1. The average grain size of the samples ECAE processed
for four passes at 425 C was of the order of 100 nm.
To the best of the authors knowledge, this is the lowest
grain size reported for a NiTiPd HTSMA. Similar
microstructures could possibly be obtained using conventional processing techniques, such as wire drawing,
however, there are no reports on the microstructures
of these dicult-to-work alloys processed using conventional techniques. In addition, an added benet of
ECAE processing is conservation of the workpiece
cross-section, which makes it possible to fabricate bulk
actuators, such as torque tubes, from NiTiPd HTSMAs.
2. The twin mode in martensite was found to be {1 1 1}
type I, and this did not change after ECAE.
3. The transformation temperatures decreased after ECAE
processing. Strengthening due to microstructural renement stabilized the thermal hysteresis, which no longer
changed with increasing constant stress levels up to
350 MPa during thermal cycling.
4. ECAE led to a signicant improvement in the thermal
cycling stability and a reduction in the irrecoverable
strain levels under constant stress.
The rationale and mechanisms responsible for these
observations are discussed in detail below.
4.1. Eect of ECAE on microstructural evolution

of 10 C min1. 150 MPa was chosen as the stress level for

the thermal cycling experiments because none of the samples
demonstrated eirr at and below 100 MPa. However, under
150 MPa load, the as-received samples started to show eirr,
while the ECAE processed sample did not exhibit a notable
one. The thermal cycling behavior of the as-received and
ECAE processed samples are shown in Fig. 11a and b up
to 10 cycles. Fig. 12a demonstrates the recovered transformation and irrecoverable strain levels as a function of number of cycles, determined from the strain vs. temperature
curves in Fig. 11. Under 150 MPa load, no eirr appeared in
any cycle and the recovered transformation strain did not
change with the number of cycles for the ECAE processed
sample, while for the as-received case, eirr obviously
increased and the recovered transformation strain level
evolved only slightly with the number of cycles.
The evolution of Ms temperature as a function of the
number of cycles was also constructed and is shown in
Fig. 12b. ECAE stabilized the Ms temperature, which is a
positive indication for the implementation of ECAE processed TiNiPd alloys in high force actuation.
4. Discussion of the results
In this work the main goal was to reveal the eect of
severe plastic deformation via ECAE and associated micro-

The TEM images and the diraction patterns in Figs. 2

and 3 showed that ECAE has no eect on the martensite
twinning mode in the Ti50.3Ni33.7Pd16 alloy. However, the
twin thickness after the processing was much smaller
than that of the as-received sample. In addition, there
was a notable renement of the grain size after ECAE.
Grain renement in binary equiatomic NiTi alloys using
ECAE triggers the formation of (0 0 1)m compound twins
with a few nanometers thickness in martensite, in addition to the common f1 1 1g type I twins [19], while the
grain renement in the Ti50.3Ni33.7Pd16 alloy did not
aect the twinning mode. The reason for such a dierence may be related to the dierent martensite structures
of NiTi and TiNiPd alloys, which are B190 and B19,
respectively. (0 0 1)m compound twinning is a lattice
invariant solution for the two-stage transformation
sequence B2 ? R ? B190 in binary NiTi [29]. It is not
known whether the same is true for the B2 ? B19
transformation.
The mechanism of grain renement during the ECAE
process should be severe plastic deformation together with
continuous dynamic recovery/recrystallization. Severe plastic deformation at elevated temperatures causes the
development of new grains as a result of a gradual increase in
misorientations between the subgrains. Further deformation

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

6419

reduces the grain size continuously and the volume fraction

of newly generated ultrane grains increases [19].

4.3. Eect of ECAE on recoverable transformation and

irrecoverable strains

4.2. Eect of ECAE on transformation temperatures

In this section the mechanism responsible for the reduction in recovered transformation and irrecoverable strains,
and the enhancement in cyclic stability after ECAE, during
the isobaric coolingheating and thermal cycling experiments will be discussed.
The recovered transformation strain levels of the
ECAE processed samples were lower at each stress level
in the isobaric coolingheating experiments. A possible
reason is the very ne grain structure in the ECAE processed samples, in which full transformation from B2 to
B19 martensite may not be attained. This argument stems
from the work by Waitz et al. [33], who investigated the
eect of ultrane grain sizes on martensitic transformation in binary NiTi. They proposed that grains smaller
than about 60 nm did not martensitically transform, even
upon large undercooling [33,34]. In addition, in order to
satisfy the compatibility across grain boundaries, the
ultrane grains in the ECAE processed samples may
not undergo full reorientation/de-twinning upon cooling.
Incomplete reorientation/de-twinning leads to a higher
volume fraction of self-accommodated, twinned martensite, limiting the achievable recoverable strain levels.
Another reason for the lower recoverable transformation
strain levels in the ECAE processed samples could be
crystallographic texture. However, since the ECAE temperatures were high, signicant texture evolution would
not be expected. Moreover, texture formation after four
passes of ECAE in several bcc and fcc materials has been
shown to be not as strong as the texture usually observed
after conventional cold working. Also, the slope of the
stresstemperature curve in Fig. 10 changed only slightly
after ECAE, indicating weak or negligible texture evolution. Therefore, crystallographic texture cannot have a
rst order eect on the recovered transformation strain
levels in the present case.
From Figs. 9a and 12a it can be observed that there
was a notable decrease in the irrecoverable strain levels
after ECAE. This was expected because, as shown in
Fig. 6, the CSS after ECAE increased as a result of
microstructural renement. On the other hand, the asreceived samples demonstrated an increase in the
irrecoverable strain levels with number of cycles in the
thermal cycling experiments, similar to the increase with
constant applied stress in the isobaric coolingheating
experiments.
An intriguing nding from the thermal cycling experiments is the enhanced cyclic stability in terms of transformation temperatures after ECAE processing. The
transformation temperatures of the ECAE processed
samples did not change with number of cycles, as presented in Fig. 12. This can also be attributed to the
increase in CSS and, thus, the suppression of additional
defect generation and a deterioration in the shape memory characteristics.

It is known that the transformation temperatures

increase and thermal hysteresis decreases upon Pd addition
to binary NiTi alloys [6,30]. However, the eect of Pd addition on the cyclic stability of NiTi as a function of thermal
cycles has not been studied in detail. It is interesting to
observe excellent thermal cycling stability of the transformation temperatures in DSC experiments of the as-received
Ti50.3Ni33.7Pd16 samples without application of any
thermo-mechanical treatment. Previous studies on equiatomic NiTi and TiNiHf alloys [10,19] have shown that
the thermal stability of transformation temperatures is only
improved after thermo-mechanical treatment in these
alloys and their thermal hysteresis was much higher than
that of the present alloy.
Recently, Delville et al. [31] showed that a TiNiPd alloy
with 11 at.% Pd possessed very low thermal hysteresis, due
to the enhanced compatibility between the single variant
martensite and austenite. Better compatibility allows twinless martensite formation and minimizes the overall energy
of the interfaces. Therefore, the improved stability of transformation temperatures and decrease in thermal hysteresis
in the present alloy, compared with binary alloys, can be
attributed to the enhanced lattice compatibility between
the austenite and martensite phases [31,32].
The transformation temperatures were decreased after
ECAE processing, as summarized in Table 1. This drop
is due to severe plastic deformation of the alloy, which
leads to grain renement and formation of dislocation substructures. In such a microstructure multiple phase fronts
should nucleate and propagate, which would require more
energy (i.e. a reduction in the transformation temperatures). Grain boundaries and high dislocation densities
act as barriers to the propagation of phase interfaces
and, thus, necessitate the nucleation of new phase fronts.
A decrease in transformation temperatures due to grain
renement has also been reported in equiatomic NiTi after
ECAE processing [19].
In summary, after ECAE the transformation temperatures decreased and thermal hysteresis increased slightly.
Note, however, that the above is only valid under
stress-free conditions. Under applied stress the situation
is slightly dierent during thermal cycling. From the isobaric coolingheating experiments it can be seen that the
transformation hysteresis increased with external stress in
the as-received material, while it was quite stable in the
ECAE processed sample. This was a consequence of
strengthening the material (as clearly shown in Fig. 6)
due to microstructural renement, such that even under
high stress levels the dislocation plasticity and/or additional defect generation was negligible. Thus, energy dissipation does not notably change with increasing stress in
the ECAE case.

6420

B. Kockar et al. / Acta Materialia 58 (2010) 64116420

5. Summary and conclusions

The present work has investigated the eect of severe
plastic deformation by ECAE on the thermal cyclic stability of a Ti50.3Ni33.7Pd16 alloy. The eects of microstructural
renement on the transformation temperatures under
stress-free conditions, recovered transformation and irrecoverable strain levels under isobaric conditions, and their
cyclic stability under a constant stress have been revealed.
In the light of the experimental results the following conclusions can be drawn from the present study:
1. Severe plastic deformation by ECAE led to a renement
of the microstructure, including grain and twin size. The
grain/subgrain sizes achieved were the lowest reported
to date in NiTiPd HTSMAs.
2. Near perfect stability of the transformation temperatures during thermal cycling under stress-free conditions
was attained in the as-received NiTiPd alloy. The stability of the transformation temperatures under constant
stress levels was enhanced after ECAE. The ECAE processed samples exhibited good thermal cyclic stability in
the constant stress thermal cycling and isobaric cooling
heating experiments. This improvement could be attributed to the increase in CSS due to microstructural
renement.
3. Microstructural renement also led to a notable
decrease in the irrecoverable strain levels in isobaric
coolingheating experiments and thermal cycling under
a constant stress.
4. The recovered transformation strain levels of the
Ti50.3Ni33.7Pd16 alloy decreased after ECAE due to
the reduced transformation volume and/or grain
boundary constraints that limit martensite reorientation/de-twinning.

Acknowledgements
This work was supported by the US Oce of Naval Research under contract no. N00014-03-M-0332 and partially
by the National Aeronautics and Space Administration under grant no. NNX07AB56A.
References
[1] Ma J, Karaman I, Noebe RD. Int Mater Rev 2010;55:257.
[2] Donkersloot HC, Van Vucht JH. J Less-Common Met 1970;20:83.
[3] Lo YC, Wu SK, Wayman CM. Scripta Metall Mater 1990;24:1571.

[4] Olier P, Brachet JC, Bechade JL, Foucher C, Guenin G. J Phys IV

1995;8:741.
[5] Thoma PE, Boehm JJ. Mater Sci Eng 1999;273275:385.
[6] Lindquist PG, Wayman CM. In: Duerig TW, Melton KN, Stockel D,
Wayman CM, editors. Engineering aspects of shape memory
alloys. Boston, MA: Butterworth; 1990.
[7] Cai W, Tanaka S, Otsuka K. Mater Sci Forum 2000;327:279.
[8] Otsuka K, Oda K, Ueno Y, Piao M, Ueki T, Horikawa H. Scripta
Metall Mater 1993;29:1355.
[9] Hornbogen E. J Mater Sci 1999;34:599.
[10] Kockar B, Karaman I, Kim JI, Chumlyakov Y. Scripta Mater
2006;54:2208.
[11] Khachin VN, Matveeva NM, Sivokha VP, Chernov DB, Yu K.
Doklady Akad Nauk SSSR 1981;257:167.
[12] Yang WS, Mikkola DE. Scripta Metall Mater 1993;28:161.
[13] Shimizu S, Xu Y, Okunishi E, Tanaka S, Otsuka K, Mitose K. Mater
Lett 1998;34:23.
[14] Golberg D, Xu Y, Murakami Y, Morito S, Otsuka K, Ueki T, et al.
Scripta Metall Mater 1994;30:1349.
[15] Golberg D, Xu Y, Murakami Y, Morito S, Otsuka K. Intermetallics
1995;3:3546.
[16] Waitz T, Antretter T, Fischer FD, Simha NK, Karnthaler HP. J
Mech Phys Solids 2007;55:41944.
[17] Waitz T, Kazykhanov V, Karnthaler HP. Acta Mater
2004;52:13747.
[18] Waitz T. Acta Mater 2005;53:227383.
[19] Kockar B, Karaman I, Kim JI, Chumlyakov YI, Sharp J, Yu C. J
Acta Mater 2008;56:363046.
[20] Kockar B, Karaman I, Kulkarni A, Chumlyakov Y, Kireeva IV. J
Nucl Mater 2007;361:298.
[21] Karaman I, Karaca HE, Maier HJ, Luo ZP. Metall Mater Trans A
2003;34:2527.
[22] Karaman I, Kulkarni AV, Luo ZP. Philos Mag 2005;85:1729.
[23] Noebe R, Padula II S, Bigelow G, Rios G, Garg A, Lerch B. In:
Smart structures and materials 2006: active materials; behavior and
mechanics, San Diego, CA, Bellingham WA: SPIE; 2006. p. 27991.
[24] Padula A, Bigelow G, Noebe R, Gaydosh D, Garg A. In: Mitchell
MR, Berg B, editors. SMST 2006: Proceedings of the international
conference on shape memory and superelastic technologies. Metals
Park, OH: ASM International; 2006.
[25] Barber RE, Dudo T, Yasskin PB, Hartwig KT. Scripta Mater
2004;51:373.
[26] Niendorf T, Canadinc D, Maier HJ, Karaman I. Metall Mater Trans
A 2007;38:1946.
[27] Niendorf T, Canadinc D, Maier HJ, Karaman I. Int J Fatigue
2008;30:426.
[28] Liu Y, McCormick PG. Acta Metall Mater 1994;42:2401.
[29] Sehitoglu H, Jun J, Zhang X, Karaman I, Chumlyakov Y, Maier HJ,
et al. Acta Mater 2001;49:3609.
[30] Liu Y, Kohl M, Okutsu K, Miyazaki S. Mater Sci Eng A
2004;378:205.
[31] Delville R, Schryvers D, Zhang Z, James D. Scripta Mater
2009;60:293.
[32] Cui J, Chu YS, Famodu OO, Furuya Y, Hattrick-Simpers J, James
RD, et al. Nature 2006;5:286.
[33] Waitz T, Antretter T, Fischer FD, Simha NK, Karnthaler HP. J
Mech Phys Solids 2007;55:419.
[34] Waitz T, Kazykhanov V, Karnthaler HP. Acta Mater 2004;52:137.