CP620, Shock Compression of Condensed Matter - 2001
edited by M. D. Furnish, N. N. Thadhani, and Y. Horie
© 2002 American Institute of Physics 0-7354-0068-7/02/$ 19.00
THE EFFECT OF ORIENTATION ON THE SPALL STRENGTH
OF THE ALUMINIUM ALLOY 7010-T6
M.R. Edwards, N.K. Bourne and J.C.F. Millett
Royal Military College of Science, Cranfield University, Shrivenham, Swindon, SN6 8LA, UK.
Abstract. Many polycrystalline alloys experience a pronounced anisotropy in their mechanical
properties. This can operate on three levels, these being at the unit cell, the microstructural level (due
to preferred orientation of the grain structure) or the meso-scale due to either phase distribution or
grain morphology. It is this latter feature in the aluminium alloy AA 7010 that has been chosen for
investigation. At quasi-static strain-rates, it has been observed that a change in fracture occurs
according to orientation, that being at 45° to the tensile axis along the rolling and long transverse
directions, and 90° to it in the short transverse direction. The same effects have been investigated
dynamically via plate impact, where the spall strength has been studied as a function of orientation and
heat treatment.
INTRODUCTION
manganese sulphide (MnS) stringers that orientate
themselves along the rolling direction. In this work,
spall strength was observed to be significantly
lower when loaded transverse to the MnS stringers
than when loaded in parallel to them.
The behaviour of aluminium alloys under shock
loading has been studied in some depth (4-7). This
is due to their low densities and (in some alloys at
least) high strengths which has encouraged their use
as light-weight armours and airframes. Possibly the
most thorough study on a single alloy was made by
Rosenberg et al.(S). Here, they showed that in the
alloy 2024 (Al+Cu+Mg), the HEL and spall
strength followed the same trends as the quasistatically measured yield strength, with the lowest
measured in the solution-treated material. They also
demonstrated that orientation also had an effect
upon the spall strength. Here the lower spall
strength was measured perpendicular to the rolling
direction when compared to that measured parallel
to it.
In this paper, we present data from the alloy
7010-T6, a high-strength airframe alloy
(Al+Zn+Mg+Cu), where we have measured the
HEL and spall strength in the short transverse and
longitudinal (rolling) directions. This work is part
The effect of orientation on the mechanical
properties of metals and alloys is well known and
has been studied extensively under quasi-static
conditions. This can occur at the unit cell level, due
to preferred orientations in the grain structure itself,
or due to the distribution of secondary phases
present within the microstructure. A more complete
discussion of such behaviour has been made by
Smallman (1). In contrast, similar measurements
made at dynamic strain-rates are nowhere near as
extensive. Gray et al. (2) investigated a cold-rolled
and annealed zirconium. Quasi-static testing
showed that peak stresses were ca. 2.5 times greater
in the through-thickness direction when compared
to the in-plane direction for the same plate. During
shock loading, results showed that the variation of
the Hugoniot Elastic Limit (HEL) were consistent
with the quasi-static measurements. Orientation was
also shown to have a significant effect upon damage
evolution, but a minimal effect on the pull-back
(spall) signal seen in recorded VISAR traces.
Similar measurements have also been made on a
eutectoid 1080 rail steel (3). Here the material is
crystallographically isotropic, but possesses
microstructural anisotropy due to the presence of
523
x
converted to stress-time by using the calibration
studies of Rosenberg et al. (9).
A 2.5 mm dural (aluminium alloy 6082-T6)
flyer was chosen as an impactor since it had a close
similarity in acoustic properties to the target. The
reflected complete releases from target and flyer
would interact in the centre of the 7010 target plate.
Impact velocities were chosen to be twice and three
times the HEL of the material, that is ca. 450 m s"1
and 895 m s"1. The velocities were measured via the
shorting of sequentially mounted pairs of pins to an
accuracy of ca. 0.1%. Specimen alignment was
better than 1 milliradian. The guns are sufficiently
accurate that control of the pressure in the breech
allowed the velocities to be repeated to ±1 ms"1.
A schematic showing the orientation of the
samples cut from the rolled plate is shown in Fig. 2.
of a wider study, where the effects of differing heat
treatments have also been investigated.
EXPERIMENTAL
Table 1. Materials Data.
Orientation
CL
(mm ju-s"1)
Longitudinal 6.24±0.03
Short
6.26±0.03
Transverse
6082-T6
6.40±0.03
Flyer
plate
^
cs
(mm us" )
3.06±0.03
p0
(g cm"3)
2.81±0.03
3.07±0.03
2.81±0.03
3.15±0.03
2.70±.01
rflte?\
nn
\^
ihort
r^a"
1
z
Longitudinal
' X
X
X
y^ NW ^\
X
_ _ I
x
•s
X
Rolling
direction
FIGURE 1. Specimen conFiguration and gauge placement.
Experiments were performed on 75 mm and 50
mm bore single stage gas guns. 50 x 50 x 5 mm3
tiles of the aluminium alloy 7010-T6 were cut from
a single hot rolled block, and ground such that they
were flat and parallel to within 5 optical fringes
over 50 mm. The hot working process gives a
characteristic pan-cake grain structure (with the
long axis of the grains in the rolling direction),
which shows dynamic recovery but not dynamic
recrystallisation.
Manganin stress gauges (MicroMeasurements
LM-SS-125CH-048) were supported on the back of
the alloy targets with 12 mm blocks of
polymethylmethacrylate (PMMA). In such a
conFiguration, the response time of the gauge is ca.
20 ns, due to the close impedance matching of the
PMMA, epoxy gauge backing and the epoxy
adhesive used to assemble the target assemblies
(see Fig. 1). Thus the response time of the gauge is
minimised and fine details in the wave profile can
be resolved. Voltage-time data from the gauges was
FIGURE 2. Orientation of plate impact specimens to the rolling
direction in rolled 7010-T6 plate.
RESULTS AND DISCUSSION
Quasi-static mechanical properties are given
below in Table 2. They are presented so that they
might act as a reference for the following
deductions from the shock-loading experiments.
Table 2. Quasi-static mechanical properties of 7010-T6
Tensile
Orientation
%
0.2%
Strength
Elongation
Proof
Stress
(MPa)
(MPa)
604
564
Longitudinal
17
516
563
12
Short Transverse
It can be seen that when tested in the
longitudinal orientation to the short transverse,
524
substituted instead. This gives corresponding values
for the HEL of 1.25 GPa and 1.06 GPa in the
longitudinal and short transverse orientations
respectively. From Table 2, it can be seen that these
differences in HEL follow the same trends as the
quasi-static proof stress. Thus it would appear that
the dependence of mechanical properties on the
orientation of 7010-T6 are also manifested in the
HEL of this material. Further, these values of the
HEL can be converted to the yield stress, 7, through
the relation,
7010-T6 displays higher strengths and elongations
then in the short transverse direction.
In Fig. 3, plate impact experiments at velocities
of 234 and 450 m s"1 are presented. In the lower
velocity impacts (ca. 234 m s"1), it can be seen that
the material does not display the characteristic
reloading signals that signify the presence of tensile
failure (spall). In the shots labelled 450 m s"1, clear
spall signals can be seen, which will be discussed
later in the paper. In both sets of traces, a clear
break in slope in the initial rising part of the trace,
that is the HEL, can be seen. Furthermore, there are
clear differences in the value of the HEL seen in the
longitudinal and short transverse orientations. These
occur at 0.39 GPa in the longitudinal orientation
and 0.33 GPa in the short transverse.
(2)
'HEL'
Where v (the Poisson's ratio) is 0.34, derived from
Table 1. This gives Y = 606 MPa in the longitudinal
orientation and 514 MPa in the short transverse.
From Table 2, this implies that 7010-T6 has a
small strain-rate sensitivity in the longitudinal
direction, but is strain-rate insensitive in the short
transverse. Thus, it would seem that the strain-rate
sensitivity of 7010-T6 is dependent on the
orientation.
In Fig. 4, gauge traces taken at the higher impact
velocities (at three times the HEL) are presented in
which clear spall signals are resolved.
1.5
1
« 0.5
4
0
0
0.5
1
Time (jus)
1.5
I,
I
2
FIGURE 3. Rear surface gauge traces from low velocity plate
impact experiments. Longitudinal direction is solid line, short
transverse direction is dotted.
•5
The stresses plotted above have been measured in
PMMA since the gauges were supported on the
back of the targets with 12 mm blocks of this
material. They have been converted to in-material
values (crx) using the well-known relation,
or =-
2ZP
!
0
0
0.5
1
Time (jus)
1.5
2
FIGURE 4. Rear surface gauge traces from high velocity plate
impact experiments. Longitudinal direction is solid line, short
transverse direction is dotted.
(1)
where OP is the stress measured in the PMMA, and
Zx and Zp are the shock impedances of the specimen
and the PMMA respectively. In practice, as all
stresses were measured at either the HEL or below
in both materials, the elastic impedances can be
Here, it can be seen that at both velocities, spall
has occurred. The measured values of the pull back
signal in each case are, at 450 m s"1, 0.31 GPa in the
longitudinal orientation and 0.45 GPa in the short
525
transverse, and at 895 m s~ l , 0.37 GPa in the
longitudinal orientation and 0.18 GPa in the short
transverse. Using the methods of equation 1, the
material values of spall strength in the two
orientations and impact conditions are given in
Table 3.
These trends cannot be explained by first order spall
theory, and further work, including microstructural
examination of soft-recovered specimens is in
progress.
REFERENCES.
Table 3. Spall strength 7010-T6______
Orientation
450 m
895 m s'1
Longitudinal
1.44
1.00
Short Transverse
1.19
0.58
1. Smallman, R.E., Modern Physical Metallurgy.
4th ed. 1985, London: Butterworths.
2. Gray, G.T., Bourne, N.K., Zocher, M.A.,
Maudlin, P.J. and Millett, J.C.F., in Shock
Compression of Condensed Matter - 1999, M.D.
Furnish, L.C. Chhabildas, and R.S. Hixson, Editors.
2000, AIP Press: Woodbury, NY. p. 509-512.
3. Gray, G.T., Lopez, M.F., Bourne, N.K., Millett,
J.C.F. and Vecchio, K.S.. in these proceedings.
(2001)
4. Ek, D.R. and Asay, J.R.,, in Shock Compression
of Condensed Matter 1985, Y.M. Gupta, Editor.
1986, Plenum: New York. p. 413-418.
5. Gilath, I., Eliezer, S., Dariel, M.P. and Kornblit,
L. Appl Phys. Lett. 52 (1988) 1207-1209.
6. Moshe, E., Eliezer, S., Dekel, E., Ludmirsky,
A., Henis, Z., Werdiger, M., Eliaz, N. and Eliezer,
D. J. Appl Phys. 83 (1998) 4004-4011.
7. Zheng, J. and Wang, Z.-P. Int. J. Solids
Structures 32 (1995) 1135-1148.
8. Rosenberg, Z., Luttwak, G., Yesherun, Y. and
Partom, Y. J. Appl. Phys. 54 (1983) 2147-2152.
9. Rosenberg, Z., Yaziv, D. and Partom, Y. J.
Appl. Phys. 51 (1980) 3702-3705.
At the lower impact velocity, the spall strengths are
similar, regardless of orientation. However, at
higher impact velocities, the spall strength is greater
in the longitudinal orientation when compared to
the short transverse. This is in agreement with the
quasi-static yield strengths shown in Table 2. It can
also be seen that in the longitudinal orientation,
spall strength increases with increasing impact
stress, whilst the opposite trend applies to the short
transverse. If true, this is contrary to accepted spall
theory, where spall strength would be expected to
be constant with impact stress. Thus it would
appear that first order spall theory is insufficient to
explain this behaviour.
CONCLUSIONS
Plate impact experiments have been performed
on the aluminium alloy 7010, in the peak-aged
condition, where the HEL and spall strength have
been determined as a function of orientation relative
to the rolling direction. The HEL was found to be
higher in the longitudinal direction than the short
transverse, following the trends of the quasi-static
yield stress. Further, conversion to 1-D stress
indicates that there is a degree of strain-rate
sensitivity in the longitudinal direction, whilst the
short transverse is strain-rate insensitive. This
implies that strain-rate sensitivity itself in 7010-T6
is orientation dependent.
The spall strengths have been shown to be
similar in both orientations at low impact stresses,
whilst at higher levels, spall strength is greater in
the longitudinal orientation. Again, this agrees with
the trends in quasi-static tensile strength. Results
also show that in the longitudinal orientation, spall
strength increases as impact stress gets higher,
whilst the opposite is true in the short transverse.
526