Note: Descriptions are shown in the official language in which they were submitted.
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
MAGNESIUM GADOLINIUM ALLOYS
This invention relates to gadolinium-containing magnesium
alloys, particularly those which possess high strength
combined with corrosion resistance, and an optimised
balance of strength and ductility. The described alloys
also have exceptional high temperature performance for
magnesium alloys. The alloys of the present invention
have been developed as extrusion alloys, but can be
rolled to produce sheets and are also suitable for
forging and machining. Although they can be cast
successfully to form billets, these alloys are not as
suitable to use as shape casting alloys in processes such
as die casting or sand casting as other magnesium alloys
due to a tendency to form cracks.
There is considerable prior art concerning the Mg-Y-Gd
system. '
The United States patent US3391034 teaches that binary
alloys of magnesium and 8 to llwts yttrium can be
produced that are age-hardenable.
It states that the ductility of these alloys is inversely
proportional to their yield strength, and that an
acceptable ductility is greater than 3-5%. It teaches
that for the magnesium yttrium system levels of yttrium
less than 8wt% do not produce sufficient mechanical
properties compared with other magnesium alloys.
The mechanical properties claimed in US3391034 are shown
in Table 1.
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
2
Table 1
Yttrium Content Yield Stress UTS Elongation
(wt o) (Mpa) (Mpa) o
8.2 303 344 3
9.0 323 374 6
10.6 335 374 5
The Russian patent SU1010880 teaches about magnesium
alloys containing yttrium and gadolinium, optionally with
zirconium. The two specific alloys discussed in the
patent specification have the mechanical properties
summarised in Table 2.
Table 2
Alloy Composition (wt%) Yield Stress UTS Elongation
(MPa) (MPa) (%)
4-6% Y, 8-10% Gd,0.3-1.0% Mn 378-390 393-442 4.4-9.8
5-6.5% Y, 3.5-5.5% Gd, 0.15-0.7% Zr 353-387 397-436 4.0-6.0
This prior art teaches that these types of manganese-
containing alloy form cracks while casting, but that this
effect is reduced by the replacement of the manganese
with zirconium. This teaching is silent regarding the
corrosion behaviour or isotropy of these alloys.
The Japanese patent JP10147830 teaches that an alloy
containing 1-<6 wt% Gd and 6-12 wt% Y produces good
strength at high temperature. Zirconium in an amount of
up to 2 wt% can also be present.
Also the Japanese patent JP9263871 teaches that an alloy
containing 0.8-5 wt% Y and 4-15 wt% Gd or Dy produces a
product that can be forged to produce an alloy of good
strength. There is however no recognition in this
document of the importance of not only the amount of each
alloying element but their respective ratios.
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
3
Using peak hardness as a measure some tests were carried
out on alloys with constant values of atomic percent rare
earths (Total Rare Earths), while varying the ratio of
yttrium plus other soluble lanthanides to gadolinium.
The results are as follows:
Melt At%Gd At% Y+ At% Ratio of Y Wt% Wt%Y+ Peak
Number other TRE + other Gd Other Hardness
soluble soluble soluble (Hv)
lanthanides lanthanides lanthanides
to Gd
DF9122 1.33 2.00 3.33 1.5 7.6 6.5 127
DF9123 0.83 2.50 3.33 3.0 4.8 8.2 110
DF9124 2.50 0.83 3.33 0.3 13.1 2.6 118
JP9263871 also discusses the addition of Ca and other
lanthanides, but we have found that the addition of Ca
and certain lanthanides is very deleterious to these
types of alloys.
The Chinese patent CN1676646 purports to teach that a
broad range of alloys containing 1-6 wt% Y, 6-15wt% Gd,
0.35-0.8 wt% Zr and 0-1.5 wt% Ca can be extruded to
produce extrudates of good strength, but there is little
specific description of the alloys of the Examples and no.
clear demonstration of the utility of the described
alloys near the limits of the claimed range.
All this prior art seems to be focussed on maximising the
strength of the alloy at the expense of its ductility,
but this latter is an equally important material
property. Furthermore there is no recognition in the
prior art of the effect of the levels of the different
alloying element on the corrosion behaviour of the
described alloys. What the present invention teaches is a
way to obtain improved ductility while also achieving
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
4
high strength levels, without sacrificing corrosion
resistance. None of this prior art recognises that when
two or more of lanthanides and yttrium are in the same
alloy, it is the specific ratio of their atomic
concentrations that is the key factor in the
effectiveness of the additions.
By selecting alloying additions within the range claimed
in this invention and controlling the isotropy of the
alloy, in addition to these improved mechanical
properties, the alloys of the present invention will
generally have corrosion rates of less than 100 mils per
year (mpy) in the industry standard ASTM B117 salt-fog
test, and preferably less than 50 mpy. Since the above
prior art does not mention the corrosion performance of
the described alloys and so it can be assumed that this
feature of the described alloys was in line with
conventional alloys, i.e. inferior to that of the alloys
of the present invention and greater than a corrosion
rate of 50 mpy.
In particular, in the academic published work by Rokhlin,
namely the book entitled "Magnesium Alloys Containing
Rare Earth Metals" Rokhlin, L L, published 2003, the
inventor of SU1010880 states that increasing the yttrium
content of magnesium alloys is detrimental to the
corrosion rate of the alloy as shown in Table 3. The
text states that this is due to the presence of Mgz4Y5
compounds which are cathodic to the solid solution.
Table 3
Yttrium Content Corrosion Rate
Wt% mg/cm2/hour Mills/years
0.5 0.025 48
3.8 0.14 268
10.5 0.36 690
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
In accordance with the present invention there is
provided a magnesium alloy consisting of:
2.0 to 5.0, preferably 2.3 to 4.6, at% in total of
gadolinium and at least one element selected from
5 the group consisting of soluble heavy lanthanides
and yttrium, wherein the ratio of the aggregate
amount of soluble heavy lanthanides and yttrium to
the amount of gadolinium is between 1.25:1 and
1.75:1, and preferably approximately 1.5:1,
from 0 up to 0.3 at% of zirconium, preferably at
least 0.03 at %, optionally with zinc, wherein when
zinc is present the amount of zinc is such that the
ratio of the weight of zinc to the weight of
zirconium is preferably less than 2:1, and more
preferably less than 0.75:1,
all other lanthanides, viz. lanthanum, cerium,
praseodymium, neodymium, promethium, samarium,
europium and ytterbium, in an aggregate amount of
less than at 0.2 at%, and preferably less than 0.1
at%,
the balance being magnesium, with any other element
being present in an amount of no more than 0.2 at%,
preferably no more than 0.1 at%, and more
preferably being present only as an incidental
impurity.
In this specification soluble heavy lanthanides are
defined as elements with atomic numbers 65 to 69
inclusive and 71. Soluble heavy lanthanides (SHL) are
those which display substantial solid solubility in
magnesium. They are terbium, dysprosium, holmium,
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
6
erbium, thulium and lutetium. These elements are
characterised by all of them having the same hexagonal
close packed metallic structure as possessed by yttrium
and magnesium, and by having a metallic radius of between
0.178nm and 0.173nm. They also exist only in a trivalent
state when oxidised, which thus distinguishes them from
elements such as europium and ytterbium which show both
tri- and bivalency and do not show any appreciable solid
solubility in magnesium. When present the aggregate
level of soluble heavy lanthanides should be greater than
0.1 at% in order ot contribute significantly to the
mechanical properties of the alloy. A particularly
preferred soluble heavy lanthanide is erbium.
It is well known that the strengthening of alloys by
precipitation hardening is a function of the amount and
type of particles that are formed. This effect is
related to both the amount of alloying elements that can
be dissolved in the matrix expressed as atomic percent
.20 and not as weight percent, and to the potential to
precipitate intermetallic particles by heat treatment.
The binary phase diagrams for the soluble heavy
lanthanides and magnesium, for yttrium and magnesium, and
for gadolinium and magnesium all show this potential.
From these phase diagrams it has been assumed to date
that the soluble heavy lanthanides, gadolinium and
yttrium will all strengthen magnesium in similar ways.
It has, however, surprisingly been found that when
gadolinium is present in a specific amount the addition
of a soluble heavy lanthanide or yttrium within a defined
range causes the formation of at least one indeterminate
ternary phase which affects the alloy's mechanical
properties. This at least one ternary phase requires a
ratio between the soluble heavy lanthanide or yttrium and
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
7
gadolinium of 3:2. Alloys having this ratio demonstrate
a better combination of mechanical properties, namely
strength, ductility and transverse properties, than can
be achieved using other combinations of amounts of the
lanthanides, yttrium and gadolinium. Significantly
improved properties can be found where the ratio is
between 1.25:1 and 1.75:1 for alloys which contain from
2.3 to 4.6 at% in total of gadolinium and at least one of
soluble heavy lanthanide or yttrium. Outside this range
either the strength and/or the ductility of the alloys
declines. This decline becomes noticeable when the total
amount of gadolinium, soluble heavy lanthanide and
yttrium is below 2.0 at% and above 5.0 at%.
In order to assist this precipitation hardening effect
a grain refining element can be added in an amount up to
its solid solubility limit in the alloy. A preferred
such element is zirconium. This can be added with
increasing amounts generally improving the alloy's yield
stress and elongation-to-failure properties. For such an
effect at least 0.03 atomic per cent of zirconium should
be present, and the maximum amount is the solid
solubility limit of Zr in the alloy which is generally at
about 0.3 atomic percent. However with both high and low
levels of zirconium corrosion resistance may decline.
The most preferred composition for a zirconium containing
alloy of the present invention is 5.5 to 6.5 wt% Y, 6.5
to 7.5 wt% Gd and 0.2 to 0.4 wt% Zr, with the remainder
being magnesium and incidental impurities. For some
alloy compositions the level of zirconium should be from
0.3 to below 0.35% by weight in order to pass the 50 mpy
salt-fog test.
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
8
It has been found that the presence of small amounts of
zinc are beneficial to the corrosion performance of the
alloys of the present invention, but that as the level of
zinc is increased the alloy's corrosion performance
deteriorates. Preferably the level of zinc should be
from 0.07 to below 0.5at%. There also appears to be a
linkage regarding the formation of different types of
precipitates when both zirconium and zinc are present in
the alloy, and it has been found that the ratio of zinc
to zirconium should not exceed 2:1, and should be
preferably less than 0.75:1.
Any lanthanide other than the required soluble heavy
lanthanide or yttrium should be present in a total amount
of less than 0.2 atomic per cent, and preferably below
0.1 at%, otherwise there is interference with the
formation of the desired at least one indeterminate
ternary phase as described above. Similarly any other
element should be present in an amount of no more than
0.2 at%, preferably no more than 0.1 at%, and more
preferably be present only at an incidental impurity
level.
The alloys of the present invention may be used for
extrusions, sheet, plate and forgings. Additionally they
may be used for parts machined and/or manufactured from
extrusions, sheet, plate or forgings.
Examples
A magnesium alloy DF8791 was produced containing 3.04 at
% in total of yttrium and gadolinium, where the yttrium
to gadolinium ratio was 1.52:1. Additionally it
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
9
contained 0.15 at% zirconium, with other elements being
at impurity levels.
Another magnesium alloy, DF8961, was produced containing
2.65 at% in total of yttrium and gadolinium, with an
yttrium to gadolinium ratio of 1.46:1. Additionally, it
contained 0.12 at% Zr and 0.08 at% Zn, with other
elements being at impurity levels.
Another magnesium alloy DF9380 was produced containing a
a 3.03 at% of a mixture of erbium, gadolinium and yttrium
with a soluble rare earth plus yttrium to gadolinium
ratio of 1.38:1. Additionally it contained 0.125 at%
zirconium.
All these alloys possessed yield stresses greater than
300MPa and elongations-to-failure greater than or equal
to 10 0 .
Three further magnesium alloys were tested, namely alloys
DF8915, DF9386 and DF8758, which had similar total levels
of yttrium and gadolinium to those of DF8961 but in
different ratios. DF8915 had a significantly higher
ratio of 3.9:1 and this produced a reduced yield stress
of only 25OMPa. DF9386 and DF8758 both had a
significantly lower ratio of 0.72:1 and 0.93:1
respectively. These low ratios had the effect of reducing
the ductility of these alloys to below 5% to levels that
are commercially unacceptable for this type of product.
A further alloy magnesium alloy DF9381 was produced
containing 2.99 at% of a mixture of ytterbium, gadolinium
and yttrium with a soluble rare earth plus yttrium to
gadolinium ratio of 1.39:1. Additionally it contained
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
0.121 at% zirconium. The ytterbium in this alloy is not a
soluble heavy lanthanide, and as a result of its addition
to the alloy the strength of the alloy was reduced to
unacceptably low levels.
5
A further set of test alloys were produced to examine the
effect of zirconium on corrosion for the alloys of the
present invention. Melts DF9382a to DF9382e all had the
same composition except for varying levels of zirconium.
10 Alloy DF9382a shows that if the material is zirconium
free (i.e. below detectable limits with standard
industrial spark emission spectroscopy) the corrosion
rate is above the acceptable level of 50 mils per year
corrosion in the standard salt fog test. Further, at
higher levels of zirconium for this alloy, DF9382b and
DF9382c also show this poor behaviour. However at levels
of zirconium between 0.03 at % (0.1 wt %) and 0.12 at %
(0.4 wt%) good corrosion performance is achieved. This is
demonstrated by DF9382d and DF9382e.
A summary of these test results is shown in Table 4, in
which some of the data has been rounded.
CA 02663605 2009-03-12
WO 2008/032087 PCT/GB2007/003491
11
~
00 00
N
0 /
U
W o h M ~? c~ tn
.r.
C'n co
a vl kr) M ~O
.^..
oo ~O O O~ O_v ~t
O
M O N M M N
00 N M O\ N N N N N
M kn vi V'
b ) V) N
a
x ~ '.. ... M
~+ cY V~ M I~ M p O~ ~ ~t ~Y d
O "O O O O\ ON O O O O O
a M N M M M M N M M M M
H
'rl
r-I o N N (~ C~l N M C~ [~-
A
E-i Q p O O O p p O p O O O
N ----- ------ ------ ------------- ------ ----- ------------- ------ ----- ---
--------- ------
\
(y ~, tt N ~t p ~ M
3 O O O p O O O O O
N .-+ 00 M M O~ tn vl =-= r, =-= =~- '-"
v~ Y hl O r~ Vr N ll~ N N ~y N ('ll N
,~' Q r+ .- O c~i - = . ~ ` ,-. ~-. r
U- - o -- -- ------ ------------- ------ ----- ------------- --=--- ----- -----
------- -----
3 h ~D h M N o0 N [~ l`~ [~ l~ l`~
rn No Wo ~"o
----- ------ ------ ------------- ------ ----- ------ ------ ------ ----- -----
------- ------
--
~ o N
0 O
\ M l~ V) V T V) 00 M M M M
00 V'~ ~!1 d' ^ '~ V~ CO 00 00 00 00
Q
---- ------ ------ ------------ ----- ----- ------------ ----- ----- ----- ----
------
a N O~ M h 00
~.+ l0 O =--~ ' ' ~O ~O ~O ~ l0
00 V-j ~ Uj
O ~ .- r= O v) ~D 00 ~-- N N N N N
'Z O\ O 00 - 00 tn 00 00 00 00 OO 00
Cl- QN M ON M h M M M M M M
-/r Q Q Q Q Q Q Q
Q Q Q Q Q
