International Journal of Plasticity 26 (2010) 189–203

Contents lists available at ScienceDirect

International Journal of Plasticity

journal homepage: www.elsevier.com/locate/ijplas

Thermo-mechanical response of Al 6061 with and without equal

channel angular pressing (ECAP)
Akhtar S. Khan *, Christopher S. Meredith
Department of Mechanical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA

a r t i c l e i n f o a b s t r a c t

Article history: The thermo-mechanical responses of Al 6061 before and after equal channel angular press-
Received 28 January 2009 ing (ECAP) at different strain rates and temperatures were measured. Al 6061 was solution
Received in final revised form 29 June 2009 heat treated before ECAP pressing at room temperature and subjected to up to three passes.
Available online 12 July 2009
After pressing, the billets were aged at 100 °C for 2 days. An as-received Al 6061-T651 was
studied similarly to investigate the differences between processed and non-processed
Keywords: specimens. The responses of ECAP material were determined at 30, 22, 125 and 250 °C,
Strengthening mechanisms
and at strain rates from 105 to 2530 s1; the 6061-T651 specimens were subjected to uni-
Rate-dependent material
Kolsky bar
axial compressive loading at 31, 22, 85, 150, 230 and 315 °C, and strain rates ranging from
Equal channel angular pressing 105 to 2200 s1. It was found that, the ECAP process increases the strength versus the
Aluminum 6061 T651 condition. Additionally, the Al 6061 ECAP is not sensitive to strain rate at room
and lower temperatures, but the sensitivity increases as the number of passes and/or tem-
perature are increased and this is the same for the non-processed material. Increasing the
number of passes increases the flow stress at room and lower temperatures, has almost no
effect at 125 °C and decreases at 250 °C. For both materials, the dynamic flow stress is
higher than the stress at quasi-static strain rates even when the quasi-static strain rate
regime is insensitive to strain rate. The Al 6061 has strong texture after one pass but stea-
dily increases as the number of passes are increased. This is the first study that reports on
the thermo-mechanical responses of ECAP and non-ECAP Al 6061 at such a wide range of
strain rates, including dynamic, and temperatures.
Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Aluminum alloys, like Al 6061, are some of the most widely used materials today which spans the entire range of indus-
tries. They are used in consumer products and military applications. The aircraft and aerospace industry uses aluminum
alloys because it is much lighter than steel and every kilogram of weight reduction results in greater fuel savings and higher
payloads. The car industry has increased its use of aluminum over the years as the price of gasoline has increased and the
need to reduce vehicle weight has been of paramount importance. Today, much of aluminum’s use is to reduce the weight of
the item being produced, but it has always been popular because it is easy to machine, cast, extrude, roll, etc. and many al-
loys are age-hardenable (like 6061).
Because of the widespread use of this alloy, it is important to understand their mechanical behavior when exposed to dif-
ferent loading conditions, strain rates and temperatures, and to be able to model the behavior and later, to predict the behav-
ior for any of these conditions. There have been many experimental studies on the mechanical behavior of this alloy at
different temperatures and strain rates, with many being performed in the 1960s. A study by Holt et al. (1967) concluded

* Corresponding author.
E-mail address: christ1@umbc.edu (A.S. Khan).

0749-6419/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijplas.2009.07.002
190 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

that Al 6061-T6 was effectively strain rate insensitive for strain rates between 0.009 and 910 s1 in compression. The dy-
namic experiments were performed on a split-Hopkinson Pressure Bar (SHPB). An earlier study (Barker et al., 1964) came
to a different conclusion when their experiments showed that Al 6061-T6 is slightly strain rate sensitive from quasi-static
to dynamic strain rate regimes. The flow stress was only slightly higher for the shock loaded material but they suggested that
a ‘‘different stress–strain curve exists for each strain rate.” They used a high speed plate impact setup and only gave an
approximate strain rate of 105 s1. Several dynamic studies on Al 6061-T6 in tension have shown that the yield strength only
increases slightly in the dynamic regime but the ultimate strength rises more significantly (20%) (Austin and Steidel, 1959;
Steidel and Makerov, 1960; Smith, 1963). Hoge (1966) observed a significant strength increase between 4.8  105 and
65 s1 that the others did not report and an increase in strain rate sensitivity with temperature. The main weaknesses of
these studies are that each only investigated a narrow range of strain rates and temperatures (or no range at all), for one
dimensional stress and strain conditions, and experiments were performed using dissimilar equipment and loading direc-
tions (i.e. tension or compression).
Equal channel angular pressing (ECAP) is a technique using severe plastic deformation to produce ultra-fine grain sizes in
the range of hundreds of nanometers to bulk course grained materials (Segal et al., 1981; Valiev et al., 1993; Segal, 1995;
Iwahashi et al., 1996; Furukawa et al., 1998). ECAP is performed by pressing a billet of material through a die that has
two channels which intersect at an angle. The billet experiences simple shear deformation, at the intersection, without
any precipitous change in the cross section area because the die does not allow for lateral expansion. This means the billet
can be pressed more than once and can be rotated about the pressing axis during subsequent pressings. A single pass with
channels 90° to each other, induces approximately 1.15 equivalent strain in the billet. Depending on the billet rotation, dif-
ferent deformation routes can be applied. Route A has no rotation of the billet, route BA is rotated counterclockwise 90° on
even number of passes and clockwise 90° on odd number of passes, route BC is rotated counterclockwise 90° after every pass,
and route C is rotated 180° after every pass (Furukawa et al., 1998). This technique can be applied to commercial pure metals
and metal alloys, with FCC, BCC and HCP crystal structures, with coarse grains to fabricate ultra-fine grained materials that
have no porosity and higher strength than the non-processed material (Valiev et al., 1997; Shin et al., 2000; Valiev et al.,
1999; Nemoto et al., 1999; Stolyarov et al., 2001; Mukai et al., 2001; Kim et al., 2002a,b). A number of high strain rate exper-
iments on ultra-fine grained and nanocrystalline materials subjected to different processing methods, in addition to ECAP,
have been performed by Gray et al. (1997), Jia et al. (2001), Wei et al. (2004), Meyer et al. (2007) and Farrokh and Khan
(2009). Constitutive modeling of UFG materials has been achieved by Farrokh and Khan (2009), Alexandrov et al. (2008),
Beyerlein and Tomé (2007), Beyerlein et al. (2007), and Muszka et al. (2006).
Al 6061 can be heat treated to greatly improve its strength and because of this, Al 6061 has been studied by other
researchers to determine pre- and post-ECAP heat treatments to optimize the increase in strength. Ferrasse et al. (1997)
studied the effects of pre- and post-pressing heat treatments of Al 6061 and concluded that after a solution heat treatment
and aging at 170 °C for 8 h before pressing (and no post-pressing heat treatment), the material exhibited peak strength and
elongation in tension. Additionally, pressing temperatures from 110 to 170 °C showed the highest strengths at the lowest
temperature. Kim et al. (2002a,b) modified the above heat treatment approach by applying only the solution treatment be-
fore pressing at 125 °C and varying the aging temperature and the time after pressing. The hardness was maximized at an
aging temperature of 100 °C for 48 h. The subsequent ultimate tensile strength and elongation values were 13% and 51%
higher than those reported by Ferrasse et al. (1997). Kim et al. (2005) investigated the cooling rate of the solution treatment
and the pressing temperature on the tensile strength of Al 6061 ECAP. Water quenching produced higher strength versus
furnace cooling, following the solution heat treatment. Interestingly, pressing at 125 °C showed higher tensile strength than
pressing at 70 °C for no post-press aging and with post-press aging, probably due to greater dynamic precipitate hardening at
the higher temperature (Roven et al., 2008). Only a couple of investigations have been published when ECAP of Al 6061 was
performed at room temperature. Horita et al. (2001) studied the strength in tension of fully annealed Al 6061 and no post-
press heat treatment. The resulting yield strength was 42% lower and the material exhibited lower elongation as compared
to Kim et al. (2002a,b). Another study at room temperature (Chang and Shan, 2003) was forced to perform intermediate
annealing at 200 °C for 1 h after each pass to prevent failure of the billet. This resulted in no strength increase from the first
to the second pass and caused a reduction in strength after two passes. Hockauf et al. (2008) successfully processed, via
ECAP, Al 6082 without cracking for up to eight passes at room temperature using backpressure. They reported a yield
strength, in tension, of 405 MPa after subjecting the billet to eight passes.
Studies on aluminum alloys have concluded that ECAP can be an effective processing method to control texture evolution
because of the many different processing parameters of the technique. Two of the most important are the number of passes
and the deformation route. Many different texture orientations with all types of texture strength can be produced at many
different numbers of passes. But the initial texture has a limited role in the outcome (Ferrasse et al., 2004). Additionally, as
the number of passes increase, the strength of the textures tends to decrease (Ferrasse et al., 2004; Wang et al., 2005; Chow-
dhury et al., 2008) due to the creation of ultra-fine grains with high angles of misorientation which limit slip (Ferrasse et al.,
2004).
This study is the first reporting the thermo-mechanical behavior of Al 6061 processed via ECAP (at room temperature)
subjected to such a wide range of temperatures and strain rates. In this study, ultra-fine grained Al 6061 specimens were
processed by ECAP for up to four passes at room temperature. The texture is observed without ECAP and at one, two, three
and four passes to determine the evolution due to the ECAP process. Further, the temperature and strain rate (quasi-static to
dynamic) sensitivities of the material under uniaxial compressive loadings were investigated to elucidate their changes as the
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 191

number of passes increases. Additionally, the mechanical behavior with ECAP is compared to Al 6061-T651 without ECAP
processing. Constitutive modeling will be investigated in a future paper.

2. Experimental procedures

2.1. Material

An age-hardenable aluminum alloy was investigated in this study, Al 6061, which is an Al–Mg–Si alloy. The Al 6061-T651
specimens were obtained from 2.54 cm round bar stock. The specimens were machined such that the specimen axis was par-
allel to the extrusion direction of the rod. These specimens were not subjected to ECAP.
The material subjected to the ECAP process was Al 6061-T6511. The specimens were obtained from 1.27 cm round bar
stock and they were machined such that the pressing axis and specimen axis were parallel to the extrusion direction of
the rod.

2.2. Equal channel angular pressing procedures

Billets 10.2 mm in diameter and 32 mm in length were solution heat treated at 530 °C for 4 h and quenched in room tem-
perature water. The ECAP die had a channel angle of 90° and a corner angle of approximately 0° (Fig. 1). For these dimen-
sions, the equivalent strain for each pass subjected to each specimen is about 1.15. Tungsten disulfide (WS2) grease was
used as a lubricant. The solution heat treated billets were pressed at room temperature for up to three passes using a press-
ing speed of 0.1 mm/s and using route BC. Four passes were attempted but the success rate was too low to report any data.
Others (Ferrasse et al., 1997; Kim et al., 2002a,b, 2005) pressed at temperatures exceeding 100 °C with higher pressing
speeds (several mm/s) because they were not able to produce samples without fracture. The pressing speed in this study
is much lower which has been shown to prevent shear band localization (Semiatin et al., 1999). When the billets come
out of the die they are assigned a billet coordinate system based on the geometry of the die. The x, y and z planes are labeled
in Fig. 1. The x, y and z directions (not drawn in Fig. 1) are perpendicular to the x, y and z planes. After pressing, the samples
were aged at 100 °C for 48 h.

2.3. Uniaxial quasi-static compression experiments

All quasi-static compression specimens were 9.5 mm in diameter and 14.2 mm in length, which maintains a length to
diameter ratio of 1.5. A MTS axial/torsional servo-hydraulic 809 system, with an axial load capacity of 250 kN, was used
to perform all of the quasi-static experiments. Strain was measured by correcting from a ‘‘blank test” which provides the

Fig. 1. Schematic of ECAP Die (Kim et al., 2002a,b).

192 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

displacement of the machine, platens, and lubrication. Strain gages were not used because most experiments were not per-
formed at room temperature. A layer of TeflonÒ tape was applied to the top and bottom platens and a film of Dow CorningÒ
lubricating grease was applied to the tape. This prevents barreling of the specimen and ensures a uniaxial stress state. Exper-
iments performed on the ECAP specimens were ‘‘jump tests” meaning three different strain rates were utilized during each
experiment. For room temperature experiments, the strain rates were 103 then 105 and finally 101 s1. The experiments
on as-received Al 6061-T651 at different strain rate were performed at 104, 102 and 100 s1 as separate experiments. The
MTS software makes it easier to program a constant displacement rate when loading the specimen, and as a result, the true
strain rate will increase in compression.
To study the material response over a wide range of temperatures, the same engineering strain rates given before were
conducted at temperatures of 30, 125 and 250 °C for the ECAP processed material and at strain rates of 103 then 104 and
finally 101 s1. The temperatures were -31, 22, 85, 150, 230 and 315 °C for the non-processed material subjected to the
same strain rates as before. Alumina rods were used between the specimen and the machine grips to reduce the heat transfer
to the machine. A layer of Dow CorningÒ, high or low temperature lubricating grease, was applied to the alumina bar surface
to prevent barreling. The specimen was maintained at the desired temperature to ensure a uniform temperature distribution
while subjected to a stress-free boundary condition. The MTS was programmed to maintain zero axial force while allowing
displacement during the heating/cooling process to allow for the thermal expansion/contraction of the alumina bars.

2.4. Uniaxial dynamic compression experiments

The split-Hopkinson pressure bar technique (SHPB) was utilized to perform dynamic uniaxial compression experiments
at engineering strain rates between 1720 and 2530 s1. In this study, the pressure bars were made out of 12.7 mm diameter
Vascomax C350 maraging steel. The high strain rate specimens were 9.5 mm in diameter and 3.8 mm in length, which is a
length to diameter ratio of 0.4. An annealed Copper C110 ‘‘pulse shaper” of 6.35 mm in diameter and 0.51 mm thickness was
used to dampen the undesirable oscillations and noise from the test. This was placed at the impact side of the incident bar.
Dynamic experiments at different temperatures (the same temperatures given before for each material) were performed for
both materials. The same Dow CorningÒ high and low temperature lubricants were used at the interfaces of the specimen
with the bars.
More detailed descriptions of the SHPB technique is given in Khan and Zhang (2000), Khan et al. (2004) and Khan et al.
(2007).

2.5. Texture analysis and surface analysis

Texture data were obtained via neutron diffraction with experiments conducted at the NIST Center for Neutron Research
for the ECAP Al 6061. Since neutrons interact weakly with matter, they easily penetrate through many millimeters of mate-
rial, thus the diffracted signal is weak, requiring long measurement times. However, the entire sample volume contributes to
the signal and gives truly average texture of the sample being measured. Additionally, every angle is accessible and the com-
plete pole figure can be obtained (Banovic et al., 2004).
The sample was a cylinder 7 mm in diameter and 7 mm in length. The samples were mounted in the center of a four circle
diffractometer with the cylindrical axis perpendicular to the plane (Fig. 1). Pole figure data were collected over the entire
hemisphere of sample orientations, which ranged from 0° to 355° in the azimuthal angle, u, and 0°–90° in the polar angle,
x, each in 5° increments. The background data were subtracted out and the intensities at each point were normalized in
multiples of random distribution (MRD). Texture data were obtained for zero, one, two and three passes.
Sample surfaces subjected to one, two and three passes of ECAP were investigated using an optical microscope with polar-
ized light. Samples were sectioned parallel to the y plane (in Fig. 1) and the surfaces were etched using Poulton’s reagent
(2 mL HF, 3 mL HCl, 20 mL HNO3, 175 mL water).

3. Results and discussion

All of the stress–strain results for all of the quasi-static, dynamic and biaxial experiments are based on the current con-
figuration of the specimen, i.e. Cauchy or true stress and true strain. Additionally, the compressive stress and strain were
assumed to have positive signs as the convention in this study.

3.1. Uniaxial compression experimental results for Al 6061-T651

Fig. 2 shows the true stress–strain response of Al 6061-T651 from uniaxial compression experiments at 31 °C and at
strain rates of 104, 102, 1 s1 and at a dynamic strain rate of 1900 s1. Figs. 3–7 are true stress–strain plots of responses
at temperatures of 22 (room temperature), 85, 150, 230 and 315 °C and at the same quasi-static and dynamic strain rates.
The dynamic strain rate at each temperature varies between 1720 and 2200 s1. This is because experiments performed at
the same hitter velocity with specimens elevated to different temperatures, and thus different strengths and elongations,
will result in different strain rates. It is difficult to adjust the hitter velocity to achieve matching strain rates from one exper-
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 193

600

500
1900 s-1 10-2 s-1

400

True Stress (MPa)

10-4 s-1 100 s-1
300

200

100

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 2. Uniaxial compression experiments at 31 °C and different strain rates for Al 6061-T651.

600

579 s-1
500
2200 s-1
True Stress (MPa)

400

10-4 s-1
10-2 s-1
300
100 s-1

200

100

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 3. Uniaxial compression experiments at 22 °C and different strain rates for Al 6061-T651.

600

500
100 s-1
1950 s-1
True Stress (MPa)

400
10-4 s-1

300
10-2 s-1

200

100

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 4. Uniaxial compression experiments at 85 °C and different strain rates for Al 6061-T651.
194 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

600

500

True Stress (MPa)

400
1720 s-1
100 s-1
10-4 s-1
300

10-2 s-1
200

100

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 5. Uniaxial compression experiments at 150 °C and different strain rates for Al 6061-T651.

400

1910 s-1
350

300
100 s-1
True Stress (MPa)

250
10-2 s-1

200
10-4 s-1
150

100

50

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 6. Uniaxial compression experiments at 230 °C and different strain rates for Al 6061-T651.

250
2110 s-1

200
True Stress (MPa)

100 s-1
150

10-2 s-1
100
10-4 s-1

50

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 7. Uniaxial compression experiments at 315 °C and different strain rates for Al 6061-T651.
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 195

600

500

True Stress (MPa)

400

300

200

-31 C,
-24F, 1900 s-1
1900/s 22 C,
70F, 1900 s-1
2200/s
100 85 C,1950/s
185F, 1950 s-1 150 C,
300F, 1950 s-1
1720/s
230 C,
450F, 1910 s-1
1910/s 315 C,
600F, 1910 s-1
2110/s
0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 8. Dynamic uniaxial compression experimental results at different temperatures for Al 6061-T651.

iment to the next in this case. The several percent difference between each dynamic strain rate is not significant because of
the many orders of magnitude difference between the strain rates at any one temperature. Fig. 8 shows a plot of only the
dynamic experiments at the different testing temperatures and shows consistent flow stress drops due to thermal softening.
At 31 °C, the quasi-static strain rate sensitivity is very low or non-existent. Interestingly, the dynamic response is mark-
edly stronger with higher work hardening, after yielding, than the quasi-static response. The yield point has changed little
between the quasi-static and dynamic responses, although SHPB data are not accurate around yield stress. Also, the quasi-
static work hardening decreases with true strain and almost reaches a plateau but the dynamic work hardening continues to
linearly increase after some initial reduction with strain, even though the thermal softening is included in the measured re-
sponse. The room temperature experiment (Fig. 3) has the same features to the response described at 31 °C but the flow
stress is a little lower at room temperature, as expected. A dynamic test was performed at 579 s1 and its strength is higher
than the quasi-static regime, which contradicts Holt et al. (1967) who reported no strain rate sensitivity up to 910 s1 and
supports the suggestion of Barker et al. (1964). The use of a pulse shaper was discussed by Franz et al. (1984) well after Mai-
den and Green, and most likely accounts for this discrepancy. The pulse shaper does two things by increasing the rise time of
the incident pulse: first it allows the sample to more easily/quickly achieve dynamic stress equilibrium. Without stress equi-
librium, the sample is not in a uniaxial stress state which is a necessary assumption to derive the stress–strain equations for
this apparatus. Secondly, the pulse shaper reduces the undesired oscillations observed in the waves, especially in the first
few percent strain. Maiden and Green do not mention a pulse shaper, but they do say that repeatability was within 10%
for the SHPB data. It is reasonable to conclude their data overlapped and appeared to be strain rate insensitive.
At 85 °C (Fig. 4), slight positive strain rate sensitivity appears after about 5% strain but there is still a large positive jump in
sensitivity from the quasi-static to the dynamic regime. The work hardening has been reduced in both strain rate regimes at
this elevated temperature, and the quasi-static response is perfectly plastic after about 15%. The behavior at temperatures of
150 °C (Fig. 5), 230 °C (Fig. 6) and 315 °C (Fig. 7) continues the same trends that developed at the lower temperatures. The
strain rate sensitivity increases, and the positive jump in sensitivity from the quasi-static to the dynamic regime decreases.
At 315 °C the strain rate sensitivity is uniform between all of the strain rates. Finally, the work hardening and strength re-
duces as the temperature increases but the dynamic work hardening is always greater than for the quasi-static regime. At
315 °C the dynamic response is perfectly plastic and the quasi-static is negative. Fig. 8 shows the dynamic response of Al
6061-T651 at different temperatures at approximately the same strain rate (average 1965 s1). The reductions in yielding
and flow stresses are consistent throughout the temperature range. The negative work hardening in Figs. 6 and 7 is due
to thermal softening of the specimens and none of the specimens failed.
The strain rate sensitivity was quantified using the equation
logðr1 =r2 Þ
m¼ ð1Þ
logðe_ 1 =e_ 2 Þ
where m is the strain rate sensitivity parameter evaluated from constant strain rate experiments, r1 j and r2 the flow stress
value at the current strain, from experiments performed with strain rates e_ 1 and e_ 2 , respectively. An average m value was
calculated by calculating separate mi values at three different strain values of 10%, 15% and 20%. Then they were averaged.
This procedure was done at each temperature and only included the quasi-static strain rates. Table 1 shows the average
strain rate sensitivity parameter at each experimental temperature. The increase of m with temperature shows quantitative
increase in the strain rate sensitivity. Almost no strain rate sensitivity is present below 85 °C, after which, the sensitivity
increases by almost an order of magnitude at the next temperature of 125 °C. Over the range of temperatures investigated,
the strain rate sensitivity increases by a factor 50.
196 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

Table 1
Comparison of the strain rate sensitivity parameter, m, at different temperatures for Al.

Temperature 31 °C 22 °C 85 °C 150 °C 230 °C 315 °C

m 0.0011 0 0.0029 0.0210 0.0378 0.0554

The flow stress can be divided into two parts – an athermal term that is mainly dependent on the long-range stress fields
due to dislocation density and grain boundaries, and the thermal term which are short-range stress fields from dislocation
interactions due to thermal activation. The athermal term is generally considered to be independent of strain rate and tem-
perature (Conrad, 1970) and will be dominant in materials with alloying elements that form precipitates. Since the alumi-
num alloys investigated here meet this criterion they are insensitive to strain rate at lower temperatures and in the quasi-
static regime. As the temperature increases, the thermal part becomes more influential, which increases the strain rate sen-
sitivity. This is due to short-range stress fields arising from dynamic precipitate hardening (Harun and McCormick, 1979). At
high strain rates, a viscous drag term is added to the flow stress and results in the observed increase in flow stress at lower
temperatures when the quasi-static regime is insensitive. At higher temperatures and strain rates, all of the stress terms are
of similar importance, and results in consistent flow stress increases as the strain rate increases.

3.2. Uniaxial compression experimental results for ECAP Al 6061

Fig. 9 shows the true stress–strain response of the Al 6061 subjected to one pass of ECAP at different post-ECAP aging
times. All of the specimens were solution heat treated and pressed in the die and aged at low temperature. An experiment
was performed on a specimen that did not undergo the ECAP processing but did have the pre- and post-ECAP heat treat-
ments applied. The single ECAP pressing increases the yield strength from approximately 125–325 MPa. The work hardening
is much less for the experiments with ECAP due to the grain size reduction which is consistent with others who have inves-
tigated nano and ultra-fine grained materials (Khan et al., 2006; Khan et al., 2008a,b). Increasing the aging time increases the
work hardening rate which saturates after 48 h, but has no effect on the yield strength. The increased work hardening is an
advantage that alloys and/or precipitate hardening alloys may have over pure metals because elevated temperatures allow
for precipitates to grow. Kim et al. (2002a,b) have observed precipitate density greatly increasing after aging at the same
temperature as this study. An aging time of 48 h is used for all subsequent experiments because it maximizes the work
hardening.
The increase in flow strength due to ECAP processing at room temperature is evident in Fig. 10. Most of the strength in-
crease occurs during the first pass with only incremental increases after subsequent passes. Additionally, the post-ECAP
aging returned most of the working hardening to the specimens after ECAP as compared to the non-pressed sample. In
the quasi-static regime, the strain rate sensitivity is non-existent. The sudden ‘‘bumps” in the stress–strain curves indicate
a change in the strain rate. These strain rate ‘‘jump” experiments were performed because there was some variability in the
flow stress between specimens and the jump experiments remove this error. After the third pass, the material displayed
slight positive quasi-static strain rate sensitivity. Kim et al. (2002a,b) has observed the precipitate size increases as the num-
ber of passes increases which is dynamic precipitate hardening. This is the mechanism for increased strain rate sensitivity in
the T651 heat treatment, but whereas in that case it was thermally activated, the severe deformation of the ECAP process is
the cause of it here. In the dynamic strain rate regime, there is an increase in strength over the quasi-static flow strength,
even though the yield strength is unchanged. Even as the quasi-static strain rate sensitivity is slight to non-existent, there
is positive strain rate sensitivity between the dynamic and quasi-static regimes. This is the same for the non-processed mate-

500
72 hrs

24 hrs
400
No Aging
True Stress (MPa)

300 48 hrs
Solution HT
No Pressing

200

100

0
0 0.05 0.1 0.15 0.2 0.25 0.3
True Strain

Fig. 9. Effect of different aging times at 100 °C subjected to one pass.

 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 197

600
2060 s-1
3 Passes
2160 s-1
500
2 Passes
2360 s-1

True Stress (MPa)

400
10-5 s-1 10-1 s-1
10-3 s-1
300

1 Pass 0 Pass, Solution HT

200

100

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 10. Quasi-static and dynamic experimental results at room temperature.

rial and is explained by the same mechanism-viscous drag. This sensitivity increases as the number of passes increases.
Meyer et al. (2007) performed quasi-static and dynamic experiments on Al 6063 and showed an increase in the rate sensi-
tivity when going from as-received to eight passes of ECAP at room temperature, similar to this study.
Fig. 11 shows the same experiments as before but at 30 °C. As expected, there is an increase in the strength of the ECAP
Al 6061 versus the room temperature experiments. There are incremental flow stress increases as the number of ECAP passes
increases. There appears to be some slight increase in the strain rate sensitivity in the quasi-static regime as the number of
passes increases but is less apparent than at room temperature. The material has the same yield strength but higher flow
stress at dynamic strain rates compared to the quasi-static regime at the same number of passes.
At an elevated temperature of 125 °C, the effect of the number of passes on Al 6061 starts an unexpected trend (Fig. 12).
Increasing the number of passes does not increase the quasi-static flow stress of the material but it does increase the strain
rate sensitivity. In addition to the increase in the experiment temperature causing an increase in the quasi-static strain rate
sensitivity, simultaneously; the increase in the number of passes causes the quasi-static strain rate sensitivity to increase.
The increase in flow strength from 104 to 101 s1 at one pass is only a few MPa, but at three passes it increases to approx-
imately 75 MPa. Additionally, as at room temperature at three passes changing the strain rate only slightly changes the
stress, at 125 °C the stress changes by many tens of MPa. Going from quasi-static to dynamic strain rates are similar to be-
fore; the yield point is the same but the flow stress is higher, but is not any more dramatic than the increase at the lower
temperatures. The dynamic flow stress does increase as the number of ECAP passes increases but the increases are smaller
than what occurred at the lower temperatures.
At experiment temperatures of 250 °C, the unexpected mechanical behavior trends were only reinforced. At this temper-
ature static recovery is occurring (Horita et al., 2001). Fig. 13 shows that in both in the quasi-static and dynamic regimes,
increasing the number of passes has become a detriment to the flow stress. This is coupled to the increase in the quasi-static
strain rate sensitivity with number of passes and with the increase in the temperature. The curves do indicate that the strain
rate sensitivity is close to reaching a plateau.

600
3 Passes 2220 s-1
2 Passes 2160 s-1
500
2230 s-1
True Stress (MPa)

400 10-1 s-1

10-4 s-1
10-3 s-1
300
1 Pass
200

100

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 11. Quasi-static and dynamic experimental results at 30 °C.

198 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

600

3 Passes 2100 s-1

500 2070 s-1
2 Passes

2230 s-1

True Stress (MPa)

400

10-1 s-1
10-3 s-1
300
10-4 s-1
1 Pass
200

100

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 12. Quasi-static and dynamic experimental results at 125 °C.

600

500

2530 s-1
True Stress (MPa)

400 3 Passes 2280 s-1

1 Pass
300 2340 s-1
10-1 s-1
2 Passes
200
10-3 s-1
10-4 s-1
100

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 13. Quasi-static and dynamic experimental results at 250 °C.

Two types of strain rate sensitivities are apparent during strain rate jump tests-instantaneous, mi, and steady-state, m,
strain rate sensitivities. mi is calculated from the instantaneous variation of the stress when the strain rate is changed from
e_ 1 to e_ 2 , and m is calculated based on the change in flow stress determined by extrapolating the steady-state flow curve back
to the moment of the change in strain rate (Picu et al., 2005). The m here is the same m given before with the as-received
material. Table 2 is a matrix of the strain rate sensitivity for different numbers of ECAP passes and at different testing tem-
peratures. The strain rate sensitivity as a function of temperature increases which is the same trend as for the as-received Al
6061-T651 before due to similar deformation mechanisms. Increasing the number of passes but at the same temperature
increases the rate sensitivity. More specifically, the as-received Al 6061-T651 was rate insensitive at room temperature
and showed slight positive sensitivity at 31 °C. This is the similar to the ECAP Al 6061, which had no sensitivity and slight
positive sensitivity for one pass and two and three passes, respectively, at room temperature. Its 30 °C behavior was also
slightly positive strain rate sensitive. The as-received Al 6061-T651 saw a big increase in the rate sensitivity at 150 °C that
was not seen after one pass of ECAP at 125 °C. This may indicate a sudden increase in rate sensitivity between these two
temperatures. Two or three passes greatly increases the rate sensitivity, showing that multiple ECAP passes can bring large
rate sensitivities to lower temperatures. As the testing temperatures increase, the ECAP Al 6061’s rate sensitivity eclipses

Table 2
Comparison of the strain rate sensitivity parameter, m, as function of the temperature and number of ECAP passes for ECAP Al 6061.

One pass Two passes Three passes

30 °C m = 0.0025 0.0022 0.0034
22 °C 0 0.00020 0.0040
125 °C 0.0037 0.0176 0.0552
250 °C 0.0517 0.0959 0.148
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 199

600

500

True Stress (MPa)

400

300

200 -30
-30 C,
C,2230/s
2230 s-1
RT, 2360/s
22 C, 2360 s-1

100 125
125 C,
C, 2230/s
2230 s-1
250
250 C,
C, 2280/s
2280 s-1

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 14. Dynamic experimental results after one pass.

that of its as-received counterpart. The as-received material has a maximum m of 0.0554 at 315 °C, which is similar to the
values of 0.0552 at 125 °C and three passes and 0.0517 at 250 °C and one pass. The increased strain rate sensitivity shifted to
lower temperatures is likely due to a lowered activation volume, and thus the spacing of short-range barriers is smaller, for
the ECAP Al 6061 compared to the as-received, which has been determined for an Al-1.5%Mg alloy subjected to ECAP (Kapoor
and Chakravartty, 2007). Vevecka-Priftaj et al. (2008) performed similar quasi-static jump experiments at different temper-
atures for Al 6061 subjected to ECAP and found the same trends, in the strain rate sensitivity as a function of temperature, as
this study. Their as-received m values were close to those in this investigation, but their ECAP processed material had greater
sensitivity (after six passes) or less sensitivity (after two passes), but the values after two passes were close to this study. The
processing parameters were different in each case, which shows the number of ECAP passes plays the dominant role in
increasing the rate sensitivity, and not (pre-) processing parameters.
Figs. 14–16 show the dynamic experiments at different temperatures for one, two and three passes. The Al 6061 ECAP
becomes more temperature sensitive as the number of passes is increased, similar to the quasi-static regime and for the
same reason.

3.3. Texture evolution and surface analysis

Fig. 17 gives the experimental {2 0 0} pole figures for the billets subjected to a different numbers of passes, with the billet
coordinate system labeled in Fig. 17(e). Fig. 17(a) shows the Al 6061 has <1 0 0> fiber texture after it was heat treated at
530 °C for 4 h but before any pressing occurs. <1 0 0> directions are roughly oriented with the z direction or the cylinder axis
and roughly axisymmetric with respect to that axis. Thus, the {1 1 1}<1 1 0> slip systems are approximately parallel to the
shearing plane (Fig. 1). After pressing once, the <1 0 0> directions are rotated away from the z direction which can partially
explain the increase in strength from the heat treated only specimen versus the specimen that was subjected to a single pass.
As the number of passes increase, the texture decreases from its maximum after one pass. By the fourth pass the texture has

600

500
True Stress (MPa)

400

300

200 -30
-30 C,
C,2160/s
2160 s-1
RT, 2160/s
22 C, 2160 s-1
100 125
125 C,
C,2100/s
2100 s-1
250
250 C,
C,2530/s
2530 s-1
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 15. Dynamic experimental results after two passes.

200 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

600

500

True Stress (MPa)

400

300

200 -30 C,
-30 C,2220/s
2220 s-1
22 C,
RT, 2060 s-1
2060/s
100 125 C,
125 C,2070/s
2070 s-1
250 C,
250 C,2340/s
2340 s-1
0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
True Strain

Fig. 16. Dynamic experimental results after three passes.

greatly reduced from the maximum even though the strength, at lower temperatures, is increased. The decrease in the tex-
ture as a function of the number of passes and observed flow stress behavior indicate that the texture formed, due to the
ECAP process, is not the dominant mechanism that changes the mechanical properties compared to non-pressed specimens.
Fig. 18 shows the sample sectioned surface parallel to the y plane after chemical etching to reveal the surface features.
Fig. 18(a) is an optical microscope image after one pass of ECAP. The arrow indicates the shear deformation direction, in
all of the images, which is approximately 45° due to ECAP processing. There is slight shear band localization that is visible
which shows that there might be benefit from increasing the pressing temperature. The shear bands are several microns
wide and are not evenly distributed across the surface. Fig. 18(b) shows well defined shear band localization, having widths
ranging from about 10 lm to about 25 lm, and showing that some thinner bands merging into larger ones. In the lighter
regions, the shear bands can be seen from the first pass. After three passes (Fig. 18(c)), the shear bands have become quite
large (about 10–50 lm), relative to the first pass, which explains why four passes results in failure of the billet. The shear

Fig. 17. Experimental {2 0 0} pole figures subjected to (a) 0, (b) 1, (c) 2, (d) 3 and (e) four passes of ECAP.
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 201

Fig. 18. Optical microscope images of samples sectioned parallel to y plane after (a) 1, (b) 2 and (c) three passes of ECAP. (The line in each figure is 100 lm).

bands are also more evenly distributed across the entire surface than from after the first pass. In all of the figures, the shear
bands are parallel to the shear strain direction, which is expected and consistent with the geometry of the ECAP die. A press-
ing temperature above room temperature or backpressure would reduce this effect and allow a higher number of passes
without macroscopic failure.

4. Conclusions

In this study, the thermal-mechanical behavior of Al 6061 subjected to the ECAP process was investigated at many dif-
ferent temperatures and strain rates, and compared to the material with no ECAP processing. It was determined that in
the quasi-static regime the strain rate sensitivity increases, from almost zero, as the number of passes were increased
and as the temperature was increased. Additionally, the flow stress increased as a function of the number of passes at room
temperature and below. At 125 °C, increasing the number of passes does not increase the flow stress (but increasing the
strain rate does increase it). At 250 °C, increasing the number of passes actually decreases the flow stress. The strain rate
sensitivity from the quasi-static to the dynamic regime is positive even when there is no quasi-static strain rate sensitivity.
The strain rate sensitivity trends as a function of temperature and strain rate regime also hold true for the as-received Al
6061-T651. The strong texture that is formed after one pass steadily declines as the ECAP passes are increased indicating
texture is not a dominant strengthening mechanism. To the authors’ knowledge, this is the first comprehensive study that
characterizes the responses at different temperatures and strain rates, including dynamic, for one material subjected to ECAP
for different numbers of passes.

Acknowledgements

This work was supported by the GAANN fellowship from the Department of Mechanical Engineering at UMBC. The
authors are grateful to Professor Horita of Kyushu University for providing the ECAP die details and acknowledge the support
of the National Institute of Standards and Technology (Dr. Gnäupel-Herold), U.S. Department of Commerce, in providing the
neutron research facilities used in this work. Muneer Baig and Amit Pandey performed a few of the experiments on Al 6061-
T651 reported in this paper.

References

Alexandrov, I., Chembarisova, R., Sitdikov, V., Kazyhanov, V., 2008. Modeling of deformation behavior of SPD nanostructured CP titanium. Mat. Sci. Eng.
A493, 170–175.
202 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203

Austin, A.L., Steidel Jr., R.F., 1959. The tensile properties of some engineering materials at high rates of strain. Proc. Am. Soc. Testing Mats., 59.
Banovic, S.W., Vaudin, M.D., Gnaeupel-Herold, T.H., Saylor, D.M., Rodbell, K.P., 2004. Studies of deformation-induced texture development in sheet materials
using diffraction techniques. Mat. Sci. Eng. A380, 155–170.
Barker, L.M., Lundergan, C.D., Herrmann, W., 1964. Dynamic response of aluminum. J. Appl. Phys. 35, 1203–1212.
Beyerlein, I.J., Tomé, C.N., 2007. Modeling transients in the mechanical response of copper due to strain path changes. Int. J. Plasticity 23, 640–664.
Beyerlein, I.J., Alexander, D.J., Tomé, C.N., 2007. Plastic anisotropy in aluminum and copper pre-strained by equal channel angular extrusion. J. Mat. Sci. 42,
1733–1750.
Chang, J.Y., Shan, A., 2003. Microstructure and mechanical properties of AlMgSi alloys after equal channel angular pressing at room temperature. Mat. Sci.
Eng. A347, 165–170.
Chowdhury, S.G., Modal, A., Gubicza, J., Krallics, G., Fodor, A., 2008. Evolution of microstructure and texture in an ultrafine-grained Al6082 alloy during
severe plastic deformation. Mat. Sci. Eng. A490, 335–342.
Conrad, H., 1970. The athermal component of the flow stress in crystalline solids. Mat. Sci. Eng. 6, 265–273.
Farrokh, B., Khan, A.S., 2009. Grain size, strain rate, and temperature dependence of flow stress in ultra-fine grained and nanocrystalline Cu and Al:
synthesis, experiment, and constitutive modeling. Int. J. Plasticity 25, 715–732.
Ferrasse, S., Segal, V.M., Hartwig, K.T., Goforth, R.E., 1997. Development of a submicrometer-grained microstructure in aluminum 6061 using equal channel
angular extrusion. J. Mat. Res. 12, 1253–1261.
Ferrasse, S., Segal, V.M., Kalidindi, S.R., Alford, F., 2004. Texture evolution during equal channel angular extrusion Part I. Effect of route, number of passes and
initial texture. Mat. Sci. Eng. A368, 28–40.
Franz, C.E., Follansbee, P.S., Wright, W.J., 1984. New experimental techniques with the split-Hopkinson pressure bar. In: Berman, I., Schroeder, J.W. (Eds.),
Eighth International Conference on High Energy Rate Fabrics, ASME.
Furukawa, M., Iwahashi, Y., Horita, Z., Nemoto, M., Langdon, T.G., 1998. The shearing characteristics associated with equal-channel angular pressing. Mat.
Sci. Eng. A257, 328–332.
Gray III, G.T., Lowe, T.C., Cady, C.M., Valiev, R.Z., Aleksandrov, I.V., 1997. Influence of strain rate & temperature on the mechanical response of ultrafine-
grained Cu, Ni, and Al-4Cu-0.5Zr. Nanostruct. Mat. 9, 477–480.
Harun, H.J., McCormick, P.G., 1979. Effect of precipitation hardening on strain rate sensitivity and yield behavior in an Al–Mg–Si alloy. Acta Metall. 27, 155–
159.
Hockauf, M., Meyer, L.W., Nickel, D., Alisch, G., Lampke, T., Wielage, B., Krüger, L., 2008. Mechanical properties and corrosion behaviour of ultrafine-grained
AA6082 produced by equal-channel angular pressing. J. Mat. Sci. 43, 7409–7417.
Hoge, K.G., 1966. Influence of strain rate on mechanical properties of 6061–T6 aluminum under uniaxial and biaxial states of stress. Exp. Mech. 6, 204–211.
Holt, D.L., Babcock, S.G., Green, S.J., Maiden, C.J., 1967. The strain rate dependence of the flow stress in some aluminum alloys. Trans. ASM 60, 152–159.
Horita, Z., Fujinami, T., Nemoto, M., Langdon, T.G., 2001. Improvement of mechanical properties for Al alloys using equal-channel angular pressing. J. Mat.
Proc. Tech. 117, 288–292.
Iwahashi, Y., Wang, J., Horita, Z., Nemoto, M., 1996. Principle of equal-channel angular pressing for the processing of ultra-fine grained materials. Scripta
Mat. 35, 143–146.
Jia, D., Ramesh, K.T., Ma, E., Lu, L., Lu, K., 2001. Compressive behavior of an electrodeposited nanostructured copper at quasi static and high strain rates.
Scripta Mat. 45 (5), 613–620.
Kapoor, R., Chkravartty, J.K., 2007. Deformation behavior of an ultrafine-grained Al–Mg alloy produced by equal channel angular pressing. Acta Mat. 55,
5408–5418.
Khan, A.S., Zhang, H., 2000. Mechanically alloyed nanocrystalline iron and copper mixture: behavior and constitutive modeling over a wide range of strain
rates. Int. J. Plasticity 16, 1477–1492.
Khan, A.S., Suh, Y.S., Kazmi, R., 2004. Quasi-static and dynamic loading responses and constitutive modeling of titanium alloys. Int. J. Plasticity 20, 2233–
2248.
Khan, A.S., Suh, Y.S., Chen, X., Takacs, L., Zhang, H.Y., 2006. Nanocrystalline aluminum and iron: Mechanical behavior at quasi-static and high strain rates,
and constitutive modeling. Int. J. Plasticity 22, 195–209.
Khan, A.S., Kazmi, R., Farrokh, B., 2007. Effect of oxygen content and microstructure on the thermo-mechanical response of three Ti-6Al-4V alloys:
experiments and modeling over a wide range of strain-rates and temperatures. Int. J. Plasticity 23, 1105–1125.
Khan, A.S., Farrokh, B., Takacs, L., 2008a. Compressive properties of Cu with different grain sizes: sub-micron to nanometer realm. J. Mat. Sci. 43, 3305–3313.
Khan, A.S., Farrokh, B., Takacs, L., 2008b. Effect of grain refinement on mechanical properties of ball-milled bulk aluminum. Mat. Sci. Eng. 489 (1–2), 77–84.
Kim, W.J., An, C.W., Kim, Y.S., Hong, S.I., 2002a. Mechanical properties and microstructures of an AZ61 Mg alloy produced by equal channel angular pressing.
Scripta Mat. 47, 39–44.
Kim, W.J., Kim, J.K., Park, T.Y., Hong, S.I., Kim, D.I., Kim, Y.S., Lee, J.D., 2002b. Enhancement of strength and superplasticity in a 6061 Al alloy processed by
equal-channel-angular-pressing. Metall. Mat. Trans. 33A, 3155–3164.
Kim, J.K., Kim, H.K., Park, J.W., Kim, W.J., 2005. Large enhancement in mechanical properties of the 6061 Al alloys after a single pressing by ECAP. Scripta
Mat. 53, 1207–1211.
Meyer, L.W., Hockauf, M., Krüger, L., Schneider, I., 2007. Compressive behavior of ultrafine-grained AA6063T6 over a wide range of strains and strain rates.
Int. J. Mat. Res. 98 (3), 191–199.
Mukai, M., Yamanoi, M., Watanabe, H., Higashi, K., 2001. Ductility enhancement in AZ31 magnesium alloy by controlling its grain structure. Scripta Mat. 45,
89–94.
Muszka, K., Hodgson, P.D., Majita, J., 2006. A physical based modeling approach for the dynamic behavior of ultrafine grained structures. J. Mat. Proc. Tech.
177, 456–460.
Nemoto, M., Horita, Z., Furukawa, M., Langdon, T.G., 1999. Microstructural evolution for superplasticity using equal-channel angular pressing. Mat. Sci.
Forum 304–306, 59–66.
Picu, R.C., Vincze, G., Ozturk, F., Gracio, J.J., Barlat, F., Maniatty, A.M., 2005. Strain rate sensitivity of the commercial aluminum alloy AA5182-O. Mat. Sci. Eng.
A390, 334–343.
Roven, H.J., Liu, M., Werenskiold, J.C., 2008. Dynamic precipitation during severe plastic deformation of an Al–Mg–Si aluminum alloy. Mat. Sci. Eng. A 483–
484, 54–58.
Segal, V.M., 1995. Materials processed by simple shear. Mat. Sci. Eng. A197, 157–164.
Segal, V.M., Reznikov, V.I., Drobyshevskiy, A.E., Kopylov, V.I., 1981. Plastic working of metals by simple shear. Russ. Metall. (Engl. Trans.) 1, 99–105.
Semiatin, S.L., Segal, V.M., Goforth, R.E., Frey, N.D., DeLo, D.P., 1999. Workability of commercial-purity titanium and 4340 steel during equal channel angular
extrusion at cold-working temperatures. Metall. Mat. Trans. 30A, 1425–1435.
Shin, D.H., Kim, B.C., Kim, Y.S., Park, K.T., 2000. Microstructural evolution in a commercial low carbon steel by equal channel angular pressing. Acta Mat. 48,
2247–2255.
Smith, J.E., 1963. Tension tests of metals at strain rates up to 200 sec1. Mat. Res. Stand. 3, 713–718.
Steidel Jr., R.F., Makerov, C.E., 1960. The tensile properties of some engineering materials at moderate rates of strain. ASTM Bull., 247.
Stolyarov, V.V., Zhu, Y.T., Lowe, T.C., Valiev, R.Z., 2001. Microstructure and properties of pure Ti processed by ECAP and cold extrusion. Mat. Sci. Eng. A303,
82–89.
Valiev, R.Z., Alexandrov, I.V., 1999. Nanostructured materials from severe plastic deformation. Nanostruct. Mat. 12, 35–40.
Valiev, R.Z., Korznikov, A.V., Mulyukov, R.R., 1993. Structure and properties of ultrafine-grained materials produced by severe plastic deformation. Mat. Sci.
Eng. A168, 141–148.
 A.S. Khan, C.S. Meredith / International Journal of Plasticity 26 (2010) 189–203 203

Valiev, R.Z., Salimonenko, D.A., Tsenev, N.K., Berbon, P.B., Langdon, T.G., 1997. Observations of high strain rate superplasticity in commercial aluminum
alloys with ultrafine grain sizes. Scripta Mater. 37, 1945–1950.
Vevecka-Priftaj, A., Böhner, A., May, J., Höppel, H.W., Göken, M., 2008. Strain rate sensitivity of ultrafine grained aluminum alloy AA6061. Mat. Sci. Forum
584–586, 741–747.
Wang, S.C., Starink, M.J., Gao, N., Xu, C., Langdon, T.G., 2005. Grain structure and texture development during ECAP of two heat-treatable Al-based alloys.
Rev. Adv. Mat. Sci. 10, 249–255.
Wei, Q., Cheng, S., Ramesh, K.T., Ma, E., 2004. Effect of nanocrystalline and ultrafine grain sizes on the strain rate sensitivity and activation volume: fcc
versus bcc metals. Mat. Sci. Eng. A 381 (1–2), 71–79.