WO2011117630A1

WO2011117630A1 - Magnesium alloy containing heavy rare earths

Info

Publication number: WO2011117630A1
Application number: PCT/GB2011/050579
Authority: WO
Inventors: Paul Lyon; Antony James Boden
Original assignee: Magnesium Elektron Limited
Priority date: 2010-03-25
Filing date: 2011-03-23
Publication date: 2011-09-29
Also published as: EP2550376A1; GB201005029D0

Abstract

Magnesium alloys which possess good processability and/or ductility whilst retaining good resistance to corrosion and/or degradation comprising one or more heavy rare earths selected from Er, Ho, Lu, Tm and Tb in a total amount of above 5.5% and no more than their respective solid solubility limit in the alloy as measured at 500°C, and Y: 0 - 10% by weight. The alloys optionally include one or more of: Nd: 0 - 5% by weight, Dy: 0 - 8% by weight, Gd:0 – 9% by weight, Zr: 0 - 1.5% by weight,Al: 0 - 10% by weight,Zn and/or Mn: 0 - 2% by weight in total, Sc: 0 - 15% by weight,In: 0 - 15% by weight, Ca: 0 - 3% by weight, and one or more rare earths and heavy rare earths other than Y, Nd, Er, Ho, Lu, Tm, Tb, Dy and Gd in a total amount of up to 0.5% by weight, the balance being magnesium and incidental impurities up to a total of 0.3% by weight.

Description

MAGNESIUM ALLOYS CONTAINING HEAVY RARE EARTHS

The present invention relates to magnesium alloys containing heavy rare earths which possess good processability and/or ductility, particularly when wrought, whilst retaining good resistance to corrosion and/or degradation, and are particularly suitable for fabrication into medical implants, for example by extrusion.

In our earlier filed International Patent application No. PCT/GB2009/002325 magnesium alloys are described which have a content of Erbium of up to 5.5% by weight and which demonstrate improvements in processability and/or ductility over known magnesium alloys such as those commercially known as Magnesium Elektron WE43 and WE54. These improved alloys also have equally good corrosion resistance of those known alloys when assessed using a standard salt fog test. Specifically for wrought applications the described alloys consist of :-

Y: 2.0 - 6.0 % by weight

Nd: 0.05 - 4.0 % by weight

Gd: 0 - 5.5 % by weight

Dy: 0 - 5.5 % by weight

Er: 0 - 5.5 % by weight

Zr: 0.05 - 1.0 % by weight

Zn + Mn:< 0.11 % by weight,

Yb: 0 - 0.02% by weight

Sm: 0-0.04% by weight,

optionally rare earths and heavy rare earths other than Y, Nd, Gd, Dy, Er, Yb and Sm in a total amount of up to 0.5 % by weight, and

the balance being magnesium and incidental impurities up to a total of 0.3 % by weight,%>, wherein

the total content of Gd, Dy and Er is in the range of 0.3- 12 % by weight, and wherein the alloy exhibits a corrosion rate as measured according to ASTM Bl 17 of less than 30 Mpy.

For casting applications the described alloys consist of

Y: 2.0 - 6.0 % by weight

Nd: 0.05- 4.0 % by weight

Gd: 0 - 5.5 % by weight

Dy: 0 - 5.5 % by weight Er: 0 - 5.5 % by weight

Zr: 0.05 - 1.0 % by weight

Zn + Mn:< 0.11 % by weight,

optionally rare earths and heavy rare earths other than Y, Nd, Gd, Dy and Er in a total amount of up to 20 % by weight, and

the balance being magnesium and incidental impurities up to a total of 0.3% by weight, wherein

the total content of Gd, Dy and Er is in the range of 0.3 - 12 % by weight, and wherein

when the alloy is in the T4 or T6 condition the area percentage of any precipitated particles having an average particle size of between 1 and 15μιη is less than 3%.

Our earlier application refers to the previous belief held by experts such as King that the behaviour of the heavy rare earths as alloying constituents was essentially the same and that therefore in magnesium alloys such as WE43 heavy rare earths such as Erbium and Ytterbium were interchangeable. Investigations revealed, however, that such prior belief was not well founded, as revealed by the solid solubility values of individual heavy rare earths in magnesium as set out in Table 2 of PCT/GB2009/002325.

Furthermore in our earlier application Gd, Dy and Er were considered to be essentially equivalent, with each being present in an amount of up to 5.5% by weight. However, the solid solubility data in Table 2 of PCT/GB2009/002325 suggested that the use of Er should be more advantageous than the use of Gd and Dy, and further work has now confirmed this. Good mechanical properties and moderate corrosion resistance has been found in alloys containing well above 5.5% Er, and whilst the tensile strength of higher content Er alloys tends to decline, especially with low levels of Y and Nd, above about 10%, ductility sufficient to enable the alloys of the present invention to be suitable for example for fabricating into medical implants has been found in some alloys containing up to 25% by weight Er.

Further work has also now established that at the temperatures that magnesium alloys are wrought Terbium is almost as soluble in magnesium as Dysprosium, Holmium possesses a solid solubility in magnesium greater than Dysprosium and is almost as soluble as Erbium, whilst Thulium, and especially Lutetium, possess solid solubilities superior to Erbium. Thus in the present invention one or more of the heavy rare earths Ho, Lu , Tm and Tb can replace the Er used in the alloys of PCT/GB2009/002325 either partly or totally.

The solid solubility limits of the heavy rare earths and selected other rare earths in pure magnesium at various temperatures, including room temperature "RT", is set out in Table 1. It will be appreciated, however, that for magnesium alloys containing other alloying elements these limits will vary.

Table 1

Although for wrought applications, particularly for structural applications, it was considered that rare earths and heavy rare earths other than Y, Nd, Gd, Dy, Er, Yb and Sm could be present in the total amount of up to 0.5% by weight, provided that the alloy exhibited a corrosion rate as measured according to ASTM Bl 17 of less than 30Mpy, it has been found that when preparing a magnesium of the type described for use as a medical implant, for example picking up the teaching of EP 141739 and 1842507 which require the alloy to be wrought, particularly by extrusion, and must meet additional criteria for such medical use, the limits as set out in PCT/GB2009/002325 can and must be adjusted. For example previously it was considered that, because of the need to retain in the alloys of PCT/2009/002325 mechanical properties, particularly tensile strength, equal to or greater than WE43 type alloys, a greater than impurity amount of cerium and lanthanum could be present and no more than an impurities amount of scandium could be tolerated. However, since for medical uses such as bone-replacement implants, such high mechanical properties are of lesser importance than the behaviour of the alloy in a biological setting, for example its biodegradable characteristics, these prior believed limitation and tolerance concerning cerium, lanthanum and scandium are in fact for the present alloys reversed. Similarly, whereas it had been thought that an alloy's corrosion behaviour in a standard salt fog test could be used as a guide to its behaviour in a corrosion test in simulated body fluid (SBF), this has been found for some alloys of the present invention not to be the case. Whilst for some their salt fog corrosion behaviour is poor, all of the alloys of the present invention have improved corrosion resistance in SBF when compared to that of the known alloys WE43 and WE54.

As explained in PCT/GB/2009/002325, it is important for good ductility that the occurrence in the alloy of large particles or clusters of particles be controlled. This is achieved by not exceeding the solid solubility limit of the rare earths and heavy rare earths, during the processing of the alloy after casting, which can involve treatment temperatures as high as 500°C.

Accordingly, the present invention provides a magnesium alloy comprising: one or more heavy rare earths selected from Er, Ho, Lu, Tm and Tb in a total amount of above 5.5% and no more than their respective solid solubility limit in the ; as measured at 500°C, wherein optionally the alloy includes one or more of:

Y: 0 - 10% by weight, Nd: 0 - 5% by weight, Dy: 0 - 8% by weight, Gd: 0 - 9% by weight, Zr: 0 - 1.5% by weight, Al: 0 - 10% by weight, Zn and/or Mn: 0 - 2% by weight in total, Sc: 0 - 15% by weight, In: 0 - 15% by weight, Ca: 0 - 3%) by weight, and one or more rare earths and heavy rare earths other than Y, Nd, Er, Ho, Lu, Tm, Tb, Dy, Gd and Er in a total amount of up to 0.5%> by weight, the balance being magnesium and incidental impurities up to a total of 0.3% by weight.

As mentioned above, magnesium has many advantages for biomedical applications, for example biodegradable inserts like stents, screws/plates/reinforcement for bone repair and surgical suture materials/staples. For many applications however, the time for degradation (corrosion) and failure of the magnesium repair device is too soon and can develop too much gas evolution (H₂) during the corrosion process. Additionally failure of stressed magnesium devices can occur due to Environmentally Assisted Cracking (EAC). EAC also referred to as Stress Corrosion Cracking (SCC) or Corrosion Fatigue (CF) is a phenomenon which can result in catastrophic failure of a material. This failure often occurs below the Yield Strength (YS). The requirement for EAC to occur is three fold, namely mechanical loading, susceptible material, and a suitable environment.

ECSS (European Cooperation for Space Standardisation), quantifies the susceptibility of various metallic alloys by use of an industry recognised test, employing aqueous NaCl solution. ECSS-Q-70-36 report ranks the susceptibility of several Magnesium alloys, including Mg-Y-Nd-HRE-Zr alloy WE54. This reference classifies materials as high, moderate or low resistance to SCC. WE54 is classed as "low resistance to SCC" ( ie poor performance). For biomedical applications, stresses are imposed on the materials and the in vivo environment (e.g. blood) is known to be most corrosive. As for the ECSS tests, SBF is widely used for in vitro testing and includes NaCl. Tests described in this patent application suggest that EAC performance of the Mg-Y-Nd-HRE-Zr alloy system can be improved by selective use of HRE additions. This offers a significant benefit for biomedical implants, where premature failure could have catastrophic results, for example Atrens (Overview of stress corrosion cracking of magnesium alloys - 8th International conference on Magnesium alloys - DGM 2009), relates potential use of stents in heart surgery, whereby fracture due to SCC would probably be fatal. The consequence of premature failure may include re intervention, patient trauma, etc. The alloys used must still be formable and show sufficient strength.

The use of the inventive Mg alloy for manufacturing an implant causes an improvement in processability, an increase in corrosion resistance and biocompatibility compared to conventional magnesium alloys, especially WE alloys such as WE43 or WE54.

The solubility of RE in magnesium varies considerably; see Table 1. It may be expected from one skilled in the art, that the volume of coarse particles present would be primarily related to the Nd content, due to the low solid solubility of this element.

Therefore the amount of RE addition may be expected to affect the amount of retained clusters and particles present in the microstructure.

Examination of the microstructure of some of the inventive magnesium alloys and conventional WE43 revealed that for specific compositions there were significantly fewer and smaller precipitates in the inventive magnesium alloy than in WE43. In other words, the selection of the type of RE, present in Mg alloy, has surprisingly led to an improvement in the formability characteristics although the total amount of RE is significantly increased. For some alloys this improvement is achieved at least in part by a reduction in hard particles (precipitates). Preferably the area percentage of any precipitated particles formed during processing of the alloy having an average particle size of between 1 and 20μιη is less than 5%. The total amount of the one or more heavy rare earths selected from Er, Ho, Lu, Tm and Tb in the Mg alloy is preferably less than 22% by weight, more preferably less than 20% by weight, preferably less than 18% by weight and more preferably less than 16% by weight. The total amount of the one or more heavy rare earths selected from Er, Ho, Lu, Tm and Tb in the Mg alloy is preferably above 6% by weight, more preferably above 6.5% by weight.

Preferably the content of at least one of the heavy rare earths selected from Er, Ho, Lu, Tm and Tb in the Mg alloy is at least 5.5% by weight.

Preferably the total content of all alloy compounds except Mg is greater than or equal to 8.5 % by weight.

The content of Y in the Mg alloy is 0 - 10.0% by weight. Preferably, the content of

Y in the Mg alloy is greater than or equal to 0.05% by weight. Preferably, the content of

Y in the Mg alloy is 0.5 - 5% by weight; most preferred 3.7 - 4.2% by weight. Keeping the content of Y within the ranges ensures that the consistency of properties, e.g. scatter during tensile testing, is maintained. Further, strength and corrosion behaviour is improved. When the content of Y is above 10.0% by weight, the ductility of the alloy deteriorates.

The content of Nd in the Mg alloy is 0 - 5% by weight, preferably greater than or equal to 0.05%> by weight, more preferably 0.05 - 2.5% by weight, and even more preferably above 1.3% by weight. When the content of Nd is above 5% by weight, the ductility of the alloy deteriorates due to a limited solubility of Nd in Mg.

Preferably the total content in the alloy of Y and Nd is at least 1% by weight so as to provide the alloy with sufficient strength for most applications. Whilst for some biomedical uses strength is of less importance than EAC behaviour it is preferred even for biomedical applications that the alloy's 0.2% YTS as measured in the as-extruded state at room temperature is at least 170 MPa.

The content of Gd in the Mg alloy is 0 - 9% by weight, preferably 0 - 7%, and more preferably no more than 5%, or for some alloys no more than 2%, by weight. Gd can reduce the degradation of the alloy in SBF tests and improve its EAC behaviour. Levels of Gd approaching the solubility limit in a given alloy reduce ductility. For some alloys Gd addition can be as low 0.2% by weight, whilst for others Gd can be avoided altogether.

The content of Dy in the Mg alloy is 0 - 8.0% by weight, preferably 0 - 4% by weight, most preferred 0 - 0.6% by weight. Dy behaves in a similar manner to Gd. The content of Er in the Mg alloy is preferably 5.5% by weight up to its solubility limit, which is in the present alloys about 25% by weight, preferably 6 - 16%> by weight, most preferred 6.5 - 11% by weight. Er can reduce the degradation of the alloy in SBF tests and improve its EAC behaviour and strength.

The content of Ho in the Mg alloy is preferably 5.5% by weight up to its solubility limit, which is in the present alloys about 25% by weight, preferably 6 - 16% by weight, most preferred 6.5 - 11% by weight. Ho can reduce the degradation of the alloy in SBF and increases strength.

The content of Lu in the Mg alloy is preferably at least 0.1 % by weight, more preferably at least 0.2% by weight. Preferably, the content of Lu in the Mg alloy is 5.5% by weight up to its solubility limit, which is in the present alloys about 25% by weight, preferably 6 - 16% by weight, most preferred 6.5 - 11% by weight. Lu can reduce the degradation of the alloy in SBF tests and improve its EAC behaviour and strength.

The content of Tm in the Mg alloy is preferably 5.5% by weight up to its solubility limit, which is in the present alloys about 25% by weight, preferably 6 - 16% by weight, most preferred 6.5 - 11% by weight. For Tm the same effect on degradation of the alloy and improvement of the EAC behaviour and strength is expected.

The content of Tb in the Mg alloy is preferably 5.5% by weight up to its solubility limit, which is in the present alloys about 25% by weight, preferably 6 - 16% by weight, most preferred 6.5 - 11% by weight. For Tb the same effect on degradation of the alloy and improvement of the EAC behaviour and strength is expected as for Er, Ho, Lu, and Tm.

A total content of Dy, Gd, Ho, Er, Lu, Tm and Tb in the Mg alloy is preferably less than 11% by weight in order to achieve a good combination of ductility, strength and EAC behaviour. In addition, the content of Zr in the Mg alloy is 0 - 1.5% by weight, preferably at least 0.01% by weight, and more preferably 0.1 - 0.9% by weight. For magnesium- zirconium alloys, zirconium has a significant benefit of reducing the grain size of magnesium alloys, especially of the pre-extruded material, which improves the ductility of the alloy. Further, Zr removes contaminants from the melt.

The content of Ca in the Mg alloy is 0 - 3% by weight, preferably 0 - 1% by weight. Ca can have a significant benefit of reducing the grain size of magnesium alloys. For some alloys Ca addition should be avoided altogether.

The content of Zn and/or Mn in the Mg alloy is 0 - 2% by weight, preferably 0 - 0.5% by weight, more preferably less than 0.40% by weight. Both Zn and Mn can contribute to precipitation and can also affect general corrosion, but Zn is preferred to Mn. For some alloys Zn addition should be avoided altogether.

The content of In in the Mg alloy is 0 % by weight up to its solubility limit, which in the present alloys can be as high as about 15% by weight, and preferably no more than 5% by weight. In can have a benefit of improving the corrosion performance of magnesium alloys. For some alloys In addition should be avoided altogether.

The content of Sc in the Mg alloy is 0% by weight up to its solubility limit, which in the present alloys can be as high as about 15% by weight, and preferably no more than 7% by weight, more preferably no more than 5% by weight. Sc can have a positive effect on corrosion resistance. For some alloys Sc addition should be avoided altogether.

Al can be added to the Mg alloy in an amount for some alloys as high as 10% by weight, but generally should be added in an amount less than 7% by weight, preferably less than 6% by weight, in order not to significantly adversely affect the alloy's strength, ductility or EAC behaviour. Preferably any Al addition should be no higher than 4% and more preferably less than 3% by weight. In some embodiments, Al addition should be less than 0.05%) by weight. For some alloys Al addition should be avoided altogether (ie Al should not be intentionally added).

The total content of impurities in the alloy should be less than 0.3% by weight, more preferred less than 0.2% by weight. In particular, the following maximum impurity levels should be preserved: Fe, Si, Cu, Mn, and Ag each less than 0.05% by weight Ni less than 0.006% by weight

La, Ce, Pr,Tb, Sm, Eu and Yb less than 0.15% by weight, preferably less than 0.1% by weight Other preferred features are set out in the accompanying claims in line with the preferred features described in PCT/GB2009/002325, such as particle size.

For purposes of the present invention, alloys are referred to as biodegradable in which degradation occurs in a physiological environment, which finally results in the entire implant or the part of the implant formed by the material losing its mechanical integrity. Artificial plasma, has been previously described according to EN ISO 10993- 15:2000 for biodegradation assays (composition NaCl 6.8 g/1. CaCl₂ 0.2 g/1. KCl 0.4 g/1. MgS0₄0.1 g/1. NaHC0₃ 2.2 g/1. Na₂HP0₄ 0.126 g/1. NaH₂P0₄ 0.026 g/1), is used as a testing medium for testing the corrosion behaviour of an alloy coming into consideration. For this purpose, a sample of the alloy to be assayed is stored in a closed sample container with a defined quantity of the testing medium at 37°C. At time intervals-tailored to the corrosion behaviour to be expected-of a few hours up to multiple months, the sample is removed and examined for corrosion traces in a known way.

DESCRIPTION OF THE DRAWINGS

The present disclosure is explained in greater detail in the following on the basis of exemplary embodiments and the associated drawings.

FIG. 1 - 7 show microstructures of samples; and

FIG. 8 shows an example of secondary cracking caused by EAC in SBF solution.

FIG. 9 shows the relative collapse pressure of stents made from WE43 and from an alloy (MI0029) according to the invention. FIG. 10 shows the relative degradation score of stents made from WE43 and from an alloy (MI0029) according to the invention.

DETAILED DESCRIPTION OF THE INVENTION Several melts with different alloy compositions were melted cast, and extruded and subsequently subject to different investigation with the emphasis on the

microstructure (grain size, size, fraction and composition of precipitates), the respective thermo-mechanical properties (tensile properties) and the corrosion behaviour with and without superimposed mechanical load. In addition bio compatibility tests were carried out. In general, melts were carried out according to the following casting technique:

High-purity starting materials (generally >99%) were melted in steel crucibles under a protective gas (C0₂/2% SF₆). The temperature was raised to 760°C to 800°C before the melt was homogenized by stirring. The melt was cast to form bars with a nominal diameter of 120 mm and a length of 300 mm. Next the bars were machined to a nominal diameter of 75 mm with a length of 150 mm to 250 mm and homogenized for 4- 8 hours. Near to the melts' solidus temperature, homogenization was typically achieved at approximately 525°C.

The material was then heated to 350-500C and extruded with the help of a hydraulic press. The resulting round rods had a diameter in the range of 6 mm to

16mm mm, mostly 9.5-12.7 mm. For the following investigations, pieces from the start and end of an extrusion 30 cm long were usually removed.

Table 2 summarises the chemical compositions, corrosion rates and tensile properties of exemplary Mg alloys. MI0007, MI0034 and SF4619 are comparative examples of WE43 within AMS4427 chemical specification used as reference material. Each time, melts were produced to generate tensile data and for metallography.

Mechanical properties and metallurgical description

To determine the mechanical properties, standardized tension tests of the bulk materials were performed and analyzed using several samples of a melt in each case. The 0.2% yield tensile strength (YTS), the ultimate tensile strength (UTS) and elongation at fracture (A) were determined as characteristic data. The yield strength YS of a material is defined as the stress at which material strain changes from elastic deformation to plastic deformation, causing it to deform permanently. The ultimate tensile strength UTS is defined as the maximum stress a material can withstand before break. In addition tensile test were also performed with extruded tubes and drawn tubes as reference. The typical extruded tubes have a typical length of not less than 30 mm, a diameter of about 2 mm and a wall thickness of between 50 and 400 μιη. They are processed by a hot micro extrusion process at temperatures between 200°C and 480°C and extrusion speeds of 0.1 mm/s to 21 mm/s. For the metallographic examination of the as extruded condition the materials were melted, cast, homogenized, cut to billets and extruded to bars. Then samples were cut, embedded in epoxy resin, ground, polished to a mirror like finish and etched according to standard metallographic techniques [G Petzow, Metallographisches, keramographisches und plastographisches Atzen, Borntraeger 2006]. Discussion of bulk material mechanical properties

Table 2 summarizes the chemical composition, mechanical (tensile test) and corrosion (salt fog in NaCl and immersion in SBF) properties of Mg alloys. As can be seen form the data of Table 2, the inventive changes in the composition of the alloys affect the tensile properties compared to the reference in terms of strength and ductility. One skilled in the art may expect increasing strength and decreasing ductility with increasing amount of appropriate alloying elements. This can actually be observed in Table 2.

For example, increasing Er content in the approximate range >2% to 8% increases strength and maintains ductility. For higher values of Er, ductility is seen to decline. Reduction in other elements, for example Nd can compensate for the Er addition, minimising the effect of Er on ductility and allowing higher additions of Er to be added; for example MI0036 shows an example of good ductility with 14 % Er addition.

Small changes in other RE additions may affect ductility - for example Gd and Dy (MI0023 vs. DF9561). Similar effects may be seen for major additions of other REs, for example Ho, Lu, and Gd however different threshold values (compared with Er) are suggested before ductility falls.

It is expected from the data, that higher levels (than the 8% examples shown) of, for example Ho and Lu, could be employed without loss of significant ductility, probably to higher values than the equivalent Er containing examples. Based on the solubility data of Table 1 similar results would be expected from the use of Tm and Tb.

As can be seen from the data of Table 2 the inventive changes of the amount of Y and Nd in the composition of the alloys basically effects strength, ductility and tolerance of some other REs.

Combinations of REs (for example Gd and Er) are not always synergistic;

however, certain combinations are expected to be beneficial.

Comparing mechanical properties of bulk material and micro-extruded tubes

Table 3: Mechanical properties of extruded bulk material and respective micro-extruded tubes

For applications, the extruded bulk material is often processed further to achieve a product. This processing can include drawing, rolling and bending steps and other advanced processing techniques. It has now been discovered that surprisingly, alloys of the invention show an improvement during such subsequent processing steps for example micro extrusion.

The comparison of the tensile properties between typical inventive alloys and the reference material before and after micro-extrusion clearly indicate that the inventive alloys are more susceptible to thermo -mechanical treatments, in particular micro- extrusion. The inventive alloy show a significant drop of 10-30% in yield strength for all tested inventive alloys, minor changes of about plus or minus 10% in ultimate strength depending on the inventive alloy and significant rise of 10 -50% in ductility for all tested inventive alloys. All of these effects are desirable because the effect of a lower YS combined with more or less the same UTS leads to a significantly lower (minus 5-30%>) yield-to-tensile strength ratio of less than 0.6 which together with the higher ductility is beneficial, for example, in terms of developing stent designs with for example homogenous opening behaviour and higher radial strength. The reference material in contrast exhibits about 20% drop in yield strength, about

10% drop in ultimate strength and about 20% drop in ductility.

Microstructure

Figures 1 through 5 show microstructures of exemplary samples (FIG. 1 : MI0031 / FIG. 2: MI0030 / FIG. 3: MI0037 / FIG. 4: MI0029 / FIG. 5: MI0046) after extrusion. They provide an insight into the effect of alloy composition upon strength and ductility of some of the alloys examples. A microstructure which is free of large particles and clusters ("clean microstructure") can offer the advantage of improved ductility if the

clusters/particles are brittle.

Figure 1 is a comparatively "clean microstructure" despite a 12.7% addition of Er - ductility is good (19%).

Figure 2 shows the effect of adding Nd to the alloy of Figure 1. The

microstructure has more clusters and ductility falls (10%). It will however be noticed that the alloy of Figure 1 possess higher tensile properties.

Figure 3 contains a higher level of Er (18%) than the alloy of Figure 1. This result in more clusters and despite an improvement in strength, ductility falls to a very low level (2%).

The alloy of Figure 4 illustrates that a combination of lower Er compared to the alloy of Figure 1 (8% Er vs. 13% Er) can achieve a comparatively "clean microstructure" and similar properties to that of alloy of Figure 1, by combining Nd with this lower Er content.

Figure 5 illustrates the effect of Lu, which appears to provide a similar manner to Er, however appears more tolerant to Nd additions in terms of freedom of particles and clusters compared with the alloy of Figure 4. Figures 6 and 7 illustrate the difference in micro-structure of drawn tubes from the reference material WE43 and micro-extruded tubes from the inventive alloy MI0029. It clearly can be seen that the micro-extruded has significantly fewer and smaller precipitates than drawn material. In addition the grain size of the extruded tubes is significantly reduced from about 15-20 μιη for the as extruded bulk materials to 2-15 μιη in the drawn condition.

In structural terms it has been found that an improvement in processability and/or ductility becomes noticeable when the area percentage of particles in the alloy having an average particle size in the range of 1 to 15 μιη is less than 3%, and particularly less than 1.5%. Most preferred the area percentage of particles having an average size greater than 1 μιη and less than 10 μιη is less than 1.5%. These detectable particles tend to be brittle.

In structural terms it has been found that an improvement in processability and/or ductility and/or strength becomes noticeable when the grain size is reduced.

Corrosion behaviour

The corrosion behaviour of selected alloy systems was investigated in greater detail on the basis of three standardized tests. The results of these tests are summarized in Table 2 and Table 4.

Identity Chemical Analysis (weight %)

Y Nd Zr Gd Dy Yb Er Sm La Ce Pr Ho Lu Al Fe TRE¹

Reference alloys

MI0034 3.8 2.3 0.54 0.44 0.47 0 0.01 0.01 0.00 - 0.01 0.002 0.9

MI0007 4 2.2 0.57 0.46 0.46 0 - - 0.00 - 0.01 0.002 0.9

SF4619 3.9 2.2 0.56 0.28 0.30 0.03 0.09 0.03 0.00 0.00 0.002 0.002 0.7

Effect of Er

DF9546 4.00 0.00 0.63 0.00 0.00 0.00 6.61 0.00 0.00 0.00 0.00 0.001 0.002 6.6

DF9561 4.00 2.20 0.61 0.00 0.00 0.00 7.14 0.02 0.00 0.00 0.00 0.01 0.003 7.2

MI0023 3.90 2.30 0.57 0.48 0.54 0.00 7.35 0.02 0.00 0.01 0.00 0.01 0.002 8.4

MI0029 4.20 2.20 0.59 0.00 0.02 0.00 7.65 0.02 0.00 0.01 0.00 0.01 0.003 7.7

MI0030 4.10 2.30 0.62 0.00 0.03 0.00 12.72 0.03 0.01 0.02 0.01 0.01 0.004 12.8

MI0036 3.60 0.03 0.83 0.00 0.00 0.00 14.00 0.01 0.00 0.01 0.00 0.01 0.004 14.0

MI0037 3.90 0.04 0.80 0.00 0.00 0.00 18.00 0.03 0.00 0.03 0.00 0.01 0.004 18.1

Effect of Y,Nd

MI0030 4.10 2.30 0.62 0.00 0.03 00.00 12.72 0.03 0.01 0.02 0.01 0.01 0.004 12.8

MI0031 3.90 0.00 0.74 0.00 0.02 0.00 12.69 0.02 0.00 0.02 0.00 0.01 0.003 12.8

MI0041 0.07 0.03 0.90 0.00 0.00 0.00 12.50 0.01 0.00 0.01 0.00 0.01 0.003 12.5

MI0042 2.1 0.21 0.80 0.00 0.01 0.00 12.50 0.01 0.00 0.01 0.00 0.01 0.003 12.5

MI0043 2.2 1.12 0.75 0.00 0.00 0.00 12.50 0.01 0.00 0.01 0.00 0.01 0.003 12.5

MI0044 1.9 1.93 0.70 0.00 0.01 0.00 12.50 0.01 0.00 0.00 0.00 0.007 0.003 12.5

Effect of Ho,Lu

MI0023 3.90 2.30 0.57 0.48 0.54 0.00 7.35 0.02 0.00 0.01 0.00 0.01 0.002 8.4

MI0045 3.7 2.07 0.73 0.43 0.29 0.00 0.03 0.06 0.00 0.03 0.00 8 0.008 0.003 8.8

MI0046 4.2 2.06 0.81 0.25 0.29 0 0.04 0.008 0 0 0.003 8 0.001 0.003 8.6

Effect of Gd & Er

MI0026 3.90 2.30 0.48 3.60 0.48 0.00 7.70 0.05 0.00 0.05 0.04 0.01 0.002 11.9

Additional Example

MI0024 4.00 2.40 0.60 0.00 0.00 0.00 1.74 0.00 0.00 0.00 0.00 0.01 0.002 1.7

Table 2 i TRE here is the sum of Gd,Dy,Yb,Er,Sm,La,Ce,Pr,Ho,Lu

Identity Tensile Properties Corrosion/Degradation Properties

0.2% YS UTS Elong NaCI SBF

MPa MPa % mpy mpy % of Reference

Reference alloys

MI0034 210 290 26 12 775

MI0007 210 291 26 14 835 100

SF4619 209 298 19 56 ND

Effect of Er

DF9546 195 283 24 12 530 40

DF9561 218 309 23 25 614 49

MI0023 276 348 14 18 206 40

MI0029 246 322 18 61 73 48

MI0030 302 370 10 74 36 23

MI0036 265 341 19 432 48 8

MI0037 302 353 2 1800 142 25

Effect of Y,Nd

MI0030 302 370 10 74 36 23

MI0031 258 337 19 NA 87 14

MI0041 159 240 24 42 252 65

MI0042 218 298 22 1588 104 27

MI0043 248 318 15 425 106 28

MI0044 248 323 18 1976 278 72

Effect of Ho,Lu

MI0023 276 348 14 18 206 40

MI0045 245 327 23 45 100 26

MI0046 228 307 24 22 66 17

Effect of Gd & Er

MI0026 299 369 11 9 276 54

Additional Example

MI0024 217 294 23 15 477 92

Table 2 (continued)

Corrosion

in SBF EAC in SBF

% of UTS in

UT WE43 SBF

Chemical Analysis wt% in type compared

SB

Reference with UTS

(Mp

ID Y Nd Zr Gd Dy Yb Er Sm La Ce Pr Ho Lu Al Fe TRE ¹ alloy in air ( % )

Reference alio; /

MI0047 4.00 2.2 0.58 0.44 0.5 0.00 0.01 0.00 0.00 0.00 0.00 0.01 0.002 1.0 100 60 243

Experimental Alloys

MI0046 4.2 2.1 0.82 0.25 0.29 0.00 0.01 0.01 0.00 0.00 0.00 8 0.01 0.002 8.6 17 70 302

MI0031 3.90 0.00 0.74 0.00 0.02 0.00 12.69 0.02 0.00 0.02 0.00 0.01 0.003 12.8 14 65 285

DF9546 4.00 0.00 0.63 0.00 0.00 0.00 6.61 0.00 0.00 0.00 0.00 0.00 0.002 6.6 40 65 268

MI0036 3.60 0.03 0.83 0.00 0.00 0.00 14.00 0.01 0.00 0.01 0.00 0.01 0.004 14.0 8 60 285

DF9561 4.00 2.20 0.61 0.00 0.00 0.00 7.14 0.02 0.00 0.00 0.00 0.01 0.003 7.2 49 60 270

MI0037 3.90 0.04 0.80 0.00 0.00 0.00 18.00 0.03 0.00 0.03 0.00 0.01 0.004 18.1 25 <45 169

MI0045 3.7 2.1 0.73 0.43 0.29 0.00 0.03 0.06 0.00 0.03 0.00 8 0.01 0.003 8.8 26 <40 <18

Additional Example

DF9400 5.50 0.00 0.35 7.19 0.00 0.00 0.02 0.05 0.00 - - 0.002 7.3 58 45 210

1 TRE here is the sum of Gd,Dy,Yb,Er,Sm,La,Ce,Pr,Ho,Lu

Table 4

Salt Fog Test

First a standardized test to evaluate the industrial usability of the alloys was performed using a 5% NaCl-containing spray mist according to ASTM Bl 17. The samples were exposed to the test conditions for the required number of days and then the corrosion product was removed by boiling in a 10% chromium trioxide solution. The weight loss of the samples was determined and expressed in mpy (mils penetration per year) as is customary in international practice.

Immersion in SBF

The corrosion resistance also depends on the corrosion medium. Therefore, an additional test method has been used to determine the corrosion behaviour under physiological conditions in view of the special use of the alloys.

For storage in SBF (simulated body fluid) with an ionic concentration of

142 mmol/L Na⁺, 5 mmol/L K⁺, 2.5 mmol/L Ca²⁺, 1 mmol/1 Mg²⁺, 1 mmol/1 S0₄ ^2~, 1 mmol/1 HP0₄ ^2~, 109 mmol/1 CI^" and 27 mmol/L HCO₃ ^" cylindrical samples of the extruded material are completely immersed in the hot medium for 7 days at nominally 37°C. The corrosion product is then removed by boiling in a 10% chromium trioxide solution. As for the ASTM Bl 17 test, the weight loss of the samples was determined and expressed in mpy.

An important factor to note is that the absolute value can vary with each batch test. This can make comparison of absolute values difficult. To resolve this, a standard (known reference WE43-type alloy, for example MI0034 type alloy) is tested with each batch of alloys tested. The reference is then used as a basis to compare any improvements.

Reference is given the value 100% and values less than this show an improvement (less degradation). However, the corrosion resistance also depends on corrosion medium and mechanical load conditions acting at the same time. Therefore, an additional test method has been used to determine the Stress Corrosion Cracking (SCC) behaviour under physiological conditions in view of the special use of the alloys. EAC/SCC in SBF

Stress tests in SBF media were carried out to identify susceptibility to

Environmental Assisted Cracking (EAC) also known as stress corrosion cracking (SCC) and compare the alloys of the invention to WE43 type reference alloy.

The test consists of testing a machined cylindrical specimen containing sharp notches to act as stress initiators. The samples were loaded with a fixed weight via a cantilever mechanism. The specimen was located inside a container which allowed SBF media to immerse the sample to a level greater than the notched portion of the sample. Media was changed every two days to minimize any compositional changes during testing. Pass criteria was at least 250 hours continuous exposure to SBF media without failure. The stress value whereby failure occurred in 250 hours or longer was defined as the threshold value which is reported in Table 4.

To determine the susceptibility of the alloy tested to the SBF media. Each batch was tested to failure in air. This value was compared with the threshold stress value in SBF as described above and the reduction in failure stress expressed as a percentage of "notched strength in air". It is likely that closer the value is to 100%, the less susceptible the material is to EAC.

Mg-Ion release from micro-extruded tubes and fully processed stents

However, since it is also known that the thermo -mechanical treatments and the surface conditions of materials affects the corrosion behaviour we also characterized the corrosion resistance of the materials by quantification of the Mg ion release from micro- extruded tubes and actual fully processed stents in SBF.

The samples for the Mg ion release tests were manufactured from micro-extruded tubes as described above. Furthermore the extruded tubes were laser beam cut to the shape of stents, electro-polished, crimped on balloon catheters, sterilized and expanded into hoses of appropriate size where they were surrounded be flowing SBF. Samples from the test solution were taken at different time points and subject to quantitative Mg ion evaluation by means of an ion chromatographic procedure described elsewhere. Drawn tubes of WE43 and the respective stent served as reference. Results of Salt fog test

Surprisingly the results of the salt fog test in NaCl atmosphere clearly indicate that an increasing content of Er (Ho, Lu, Tm, Tb) reduces the corrosion resistance

significantly. The more surprising is the oppositional corrosion behaviour of the inventive alloys in SBF a solution simulating the actual biological service environment of vascular implants much better than the salt fog test since previous investigations indicated distinct correlation between mass loss in salt fog (ASTM Bl 17) and mass loss in SBF. The SBF immersion test revealed a significant reduction of the mass loss with increasing content of Er from 6 wt% to 14 wt%. From about 18 wt% Er on the corrosion rate may begin to deteriorate.

Results of Immersion in SBF

Tests in immersed SBF, of the alloys of the invention, illustrates that the reduction on degradation rate (corrosion). This is best viewed as a percentage of the reference alloy WE43 type. In the best case examples from the invention show a greater than 10 fold improvement in degradation.

Generally speaking, as Er, Ho, Lu, Gd and, it is expected, Tm and Tb increase, the degradation resistance also improves, i.e. the measured loss becomes lower compared to the reference. Within the alloy additions mentioned above, there also appears to be some differences between the elements individually, with some showing better performance. It would be expected that combinations of some of the HREs could provide synergistic benefits at appropriate alloy contents.

Results of EAC/SCC in SBF

Table 4 provides data on EAC tests. Taking WE43 type alloy (MI0047) as a reference, it can be seen that as the HRE content increase, the absolute tolerable stress increase. This improvement is also seen as a percentage of the actual strength of the material when tested in air (no SBF media effect). The closer this value is to 100%, the less the fracture is related to the media and therefore the less prone the material is likely to be to EAC (SCC) in that media. Er additions perform well to at least 14wt %, however at 18 wt % the performance is reduced to beneath that of the reference WE43 type alloy. Other HREs perform in different ways. For example whilst Gd can improve EAC resistance of the alloys tested, and Lu also appears good, Ho performs poorly. Figure 8 shows the fracture appearance of comparative alloy DF9400. The fracture shows primary and secondary cracking. This type of cracking with secondary cracking remote from the primary crack can be representative of SCC.

Discussion: Mg-Ion release from micro-extruded tubes and fully processed stents

Table 5 Table 5 shows a comparison of the Mg ion release from bulk material, extruded tubes and the respective stents from these extruded sleeves. Values are given in percentage of the respective reference material (reference WE 43 bulk material from Table 2 as reference for the inventive bulk material, drawn tube of WE43 for the extruded tubes of the inventive alloy, and stents from drawn tube of WE43 for the stents manufactured from extruded tubes of the inventive alloys).

The Mg ion release tests with micro-extruded tubes and the respective stents as well as with drawn tube and the respective stents as the respective reference revealed that the inventive alloys exhibit significantly less Mg ion release (20 - 80% less Mg ion release than the reference material) indicating a significant higher corrosion resistance. Furthermore the inventive alloys (and the reference) become about 20% to 80% more corrosion resistant than the drawn reference material when processes by micro-extrusion. Further improvements (50 - 90% less Mg ion release than stents from drawn reference material) can be gained through proper electro-polishing to produce a stent. While the inventive alloy shows improved corrosion properties after processing, the corrosion properties of the reference material drop during tube drawing and polishing.

This can be explained by the microstructure of the bulk materials (see section "Mechanical properties and metallurgical description of the alloy") resulting from the inventive changes of the composition and the processed material where the drawn reference tube shows significantly more precipitates than the micro-extruded material. These precipitates (known to be electrochemical more noble) also exist on the polished surface creating galvanic couples that accelerate dissolution rates.

In addition the grain size is significantly reduced from about 15-20 μιη in the as extruded and drawn condition to 2-15 μιη in the micro-extruded condition. Although described with particular reference to medical implants the alloys of the present invention have a variety of other uses and can be cast and/or heat treated and/or wrought and/or used as a base alloy for a metal matrix composite and/or rapidly solidified to produce an amorphous form, such as a powder or wire, and/or extruded in any way.

Example - Production and corrosion fatigue testing of a stent using Mg-4Y-2Nd- 8Er-0.6Zr (MI0029)

High purity (>99.9%) magnesium ingots were smelted in steel crucibles at 700- 800°C. The melt was protected from burning and sludge formation using fluxless techniques with mixtures of protective gases, e.g. C0₂/2% SF₆ or argon/2% SF₆. After smelting the pure magnesium ingots, the temperature was raised to 680-860°C and the respective amounts of alloy ingredients Y, Nd and Er and Zr were added.

The melt was homogenised by stirring, and then cast in a mold to form bars with a nominal diameter of 120mm and a length of 300mm. After casting and cooling, the bars were machined to a nominal diameter of 75mm with a length of 250mm and homogenised for 8 hours at approximately 525°C. The material was then reheated to 400-500°C and extruded using a hydraulic press. The resulting round rods had a diameter of 12.7 mm. Before further processing or testing, 30 cm long pieces were removed from the start and end of the extrusions. The mechanical properties of the extruded bulk material compared to WE43 were as follows:

YTS = 246 MPa, which is ca. 35 MPa higher than for WE43 UTS = 322 MPa, which is ca. 30 MPa higher than for WE43

A = 18%, which is ca. 1-7% points less than for WE43.

For medical applications such as vessel scaffolds (also known stents) in the vascular field, the extruded material must be further processed into tubes, e.g. by drawing or micro-extrusion. Stents are endoluminal endoprostheses having a carrier structure. The structure comprises a hollow body which is open at both ends and a peripheral wall which is formed by a plurality of struts connected together. The struts can be folded in a zig-zag configuration. The struts have typical dimensions in width and thickness of 30-450μιη.

The further processing of the extruded alloy into tubes was accomplished by a micro-extrusion process. Slugs of alloy were machined from the bulk material. These slugs were processed by hot pressing at elevated temperatures of between 200°C and 480°C and extrusion speeds of 0.001 mm/s to 600 mm/s. Typically, the micro-extruded tubes for vessel scaffolds had a length of not less than 30 mm, a diameter of ca. 2 mm and a wall thickness between 50 and 400 μιη.

The mechanical properties of the micro-extruded tubes compared to the drawn WE43 tubes were as follows:

YTS = 189 MPa, which is ca. 25 MPa higher than for drawn WE43 tubes

UTS = 316 MPa, which is ca. 66 MPa higher than for drawn WE43 tubes

A = 26%o, which is ca. 6%> higher than drawn WE43 tubes

To evaluate susceptibility to environmental assisted cracking (ie corrosion fatigue) upon an in vz^'vo-like cyclic loaded vascular scaffold, stents were produced from the micro- extruded tube by laser cutting and electro-polishing.

Prior to testing, the stents were crimped onto balloon catheters to a diameter of less than 1.5 mm and sterilized, e.g. by ETO or e-beam. The stents were than over- expanded to their nominal diameter plus 0.5 mm into mock arteries with respective diameters which has been filled with simulated body fluid (SBF). Previous tests have shown that over-expansion to about 1 mm in diameter is possible for the alloy according to the invention, whereas an identical stent manufactured from WE43 tolerates significant less over-expansion. The improved dilatation reserve of the inventive alloy contributes significantly to device safety in clinical practice.

The mock arteries, with the stent inside, were placed in a test chamber where a cyclic physiological load was applied. After specific periods of time (14 and 28 days), some arteries were transferred into another test chamber where the radial strength of the stent was measured. Other arteries under test were filled with epoxy resin for

metallographic determination of the remaining load bearing cross section of the stent struts. For comparison, identical tests were performed on stents manufactured from WE43 tubing. The results of these tests are shown in Figures 9 and 10.

Figure 9 shows the relative collapse pressure over a 28 day period of a stent made from WE43 alloy and one made from the alloy of the invention. The relative collapse pressure is calculated by measuring the absolute collapse pressure and then expressing this as a percentage of the initial collapse pressure of the stent made from WE43. As depicted in Figure 9, these results showed that the stent made from the alloy of the invention exhibited a significant higher initial relative collapse pressures (+10%) as a result of greater strength, lower yield ratio and higher strain hardening. Furthermore, the stent made from the alloy of the invention maintained a high relative collapse pressure over a significant longer period of time without fractures or fragmentation, indicating a significantly lower susceptibility to environmental assisted cracking (ie corrosion fatigue).

Figure 10 shows the relative degradation score over a 28 day period of a stent made from WE43 alloy and one made from the alloy of the invention. The relative degradation score is calculated by measuring the weight of the stent and then expressing this as a percentage of the initial weight of the stent made from WE43. As shown in Figure 10, these results showed that the stent made from the alloy of the invention exhibited a significantly lower levels of corrosion (-25%) than the stent made from WE43 when under cyclic load in a corrosive environment.

Claims

CLAIMS:

1 . A magnesium alloy comprising: one or more heavy rare earths selected from Er, Ho, Lu, Tm and Tb in a total amount of above 5.5% and no more than their respective solid solubility limit in the alloy as measured at 500°C, and

Y: 0 - 10% by weight, wherein optionally the alloy includes one or more of:

Nd: 0 - 5% by weight, Dy: 0 - 8% by weight,

Gd: 0 - 9% by weight,

Zr: 0 - 1 .5% by weight,

Al: 0 - 10% by weight,

Zn and/or Mn: 0 - 2% by weight in total, Sc: 0 - 15% by weight,

In: 0 - 15% by weight,

Ca: 0 - 3% by weight, and one or more rare earths and heavy rare earths other than Y, Nd, Er, Ho, Lu, Tm, Tb, Dy and Gd in a total amount of up to 0.5% by weight, the balance being magnesium and incidental impurities up to a total of

0.3% by weight.

2. An alloy as claimed in claim 1 wherein the total of Y + Nd is at least 1 % by weight,

3. An alloy as claimed in claim 1 or claim 2 wherein the alloy has a ductility sufficient to exceed 7% elongation at fracture when measured in its as- extruded state at room temperature according to the ductility test conditions set out in ASTM 557M

4. An alloy as claimed in any one of claims 1 to 3 wherein the alloy has a corrosion resistance of better than 90% of that of the commercial alloy Elektron WE43 when measured under identical conditions in Simulated Body Fluid.

5. An alloy as claimed in any one of the preceding claims wherein the content of Y is 0.5 - 5% by weight.

6. An alloy as claimed in claim 5 wherein the content of Y is 3.7 - 4.2% by weight.

7. An alloy as claimed in in any one of the preceding claims wherein the content of Nd is 0.05 - 2.5% by weight.

8. An alloy as claimed in any one of the preceding claims wherein the total amount of the selected heavy rare earths Er, Ho, Lu, Tm and Tb is 6 - 16% by weight.

9. An alloy as claimed in claim 8 wherein the total amount of the selected heavy rare earths Er, Ho, Lu, Tm and Tb is 6.5 - 1 1 % by weight.

10. An alloy as claimed in any one of the preceding claims wherein the selected heavy rare earth is Er, Lu or Tm.

1 1 . An alloy as claimed in any one of the preceding claims wherein the content of Dy is 0 - 4% by weight.

12. An alloy as claimed in claim 1 1 wherein the content of Dy is 0 - 0.6% by weight.

13. An alloy as claimed in any one of the preceding claims wherein the content of Gd is 0 - 5% by weight.

14. An alloy as claimed in claim 13 wherein the content of Gd is 0 - 2% by weight.

15. An alloy as claimed in any one of the preceding claims wherein the total content of Er, Ho, Lu, Tm, Tb, Dy and Gd is less than 1 1 % by weight.

16. An alloy as claimed in any one of the preceding claims wherein the content of Zr is 0.1 - 0.9% by weight.

17. An alloy as claimed in any one of the preceding claims wherein the content of Al is 0 - 3% by weight.

18. An alloy as claimed in any one of the preceding claims wherein the content of Zn + Mn is 0 - 0.40% by weight. 19. An alloy as claimed in any one of the preceding claims wherein the content of Sc is 0 - 5% by weight.

20. An alloy as claimed in any one of the preceding claims wherein the content of In is 0 - 5% by weight.

21 . An alloy as claimed in any one of the preceding claims wherein the content of Ca is 0 - 1 % by weight.

22. An alloy as claimed in any one of the preceding claims having the ductility sufficient to exceed 15% elongation as defined in claim 3.

23. An alloy as claimed in any one of the preceding claims having a corrosion resistance of better than 40% as defined in claim 4. 24. An alloy as claimed in any one of the preceding claims having a magnesium content of at least 81 % by weight.

25 An alloy as claimed in any one of the preceding claims wherein when the alloy is in the T4 or T6 condition the area percentage of any precipitated particles having an average particle size of between 1 and 15μηη is less than 3%. 26. An alloy as claimed in any one of the preceding claims when cast and/or heat treated and/or wrought and/or used as a base alloy for a metal matrix composite and/or rapidly solidified to produce an amorphous form and/or A magnesium alloy as claimed in claim 1 substantially as hereinbefore

28. A magnesium alloy as claimed in claim 1 substantially as hereinbefore described in any one of the specific Examples.