EP2553131A1

EP2553131A1 - High temperature stable aluminium alloy

Info

Publication number: EP2553131A1
Application number: EP11763104A
Authority: EP
Inventors: Calin Daniel Marioara; Sigmund Jarle Andersen; Sverre Gulbrandsen-Dahl; Jon Holmestad; Randi Holmestad; Tor-Erik Nicolaisen; Inge-Erland Opheim; Oddvin Reiso; Jostein RØYSET
Original assignee: Norsk Hydro ASA
Current assignee: Norsk Hydro ASA
Priority date: 2010-03-30
Filing date: 2011-03-30
Publication date: 2013-02-06
Anticipated expiration: 2031-03-30
Also published as: WO2011122958A1; EP2553131B1; EP2553131A4

Abstract

The present invention relates to Al-Mg-Si-Cu alloy optimised for high temperature stability The alloy is characterized in that its content of Mg and Si lies within a polygon defined by the following coordinates of an Mg-Si diagram: a1 - a2 - a3 - a4 -a1 where in wt.% a1 = 0.60Mg,0.60Si, a2 = 0.90Mg, 0.90Si, a3 = 1.30 Mg, 0.60 Si and a4 = LOOMg, 0.30Si, and with the additional alloying elements: - Cu between 0.20 and 0.50 wt.% - Fe between 0.08 and 0.40 wt.%, and where at least one of the following elements are added for the purpose of grain structure control during processing of the alloy - Mn between 0 and 0.80 wt.% - Cr between 0 and 0.30 wt.% - Zr between 0 and 0.30 wt.%, and optionally Ti up to 0, 1 wt% and B up to 0,1 wt% as grain refining elements, and further optionally Ge between 0 and 0.20 wt.% and Ag between 0 and 0.20 wt.%, rest Al, including incidental impurities. In the alloy as defined above the L-phase is the dominant precipitate type as regards number density upon over-ageing.

Description

High temperature stable aluminium alloy

The present invention relates to Al-Mg-Si-Cu alloy optimised for high temperature stability.

Alloys of the Al-Mg-Si system have an attractive combination of processability, mechanical strength and response to surface finishing treatments, thus making them the alloys of choice for several applications within the building industry and the automotive industry. Some of the automotive applications require that the alloy should maintain a certain mechanical strength after defined thermal exposures. Requirements for thermal exposure are expected to increase in the future. There is therefore a strong drive in the aluminium industry for developing aluminium alloys that can meet the current and the future requirements.

A common aluminium alloy for structural applications in the European market is the standard 6082 alloy. This alloy contains sufficient amounts of Mg and Si for obtaining tensile strength of typically 300-320 MPa under common industrial practices. In addition, the alloy contains a substantial amount of Mn. The Mn forms dispersoids during homogenisation of the alloy. The purpose of the dispersoids is to control the microstructure during thermo-mechanical processing, such as for example obtaining a fibrous grain structure after extrusion of the alloy. The 6082 alloys do, however, lose strength after prolonged temperature exposure at for instance 200°C. With the present invention is provided a novel Al-Mg-Si-Cu alloy optimised for high temperature stability.

The alloy is characterized by the features as defined in the independent claim 1. Preferred embodiments of the invention are further defined in the dependent claims 2 - 8. The invention will now be further described by way of examples and with reference to the figures where:

is a diagram showing the development in strength of tested precipitation strengthened alloys 1 , 2 and 3, defined in Table 1 in the description, as a result of exposure to a temperature of 200°C, is Transmission Electron Microscopy images of precipitate types of a) alloy 1 , b) alloy 2 and c) alloy 3 after exposure of temperature of 200°C for 7 days. is a diagram showing the development in strength of precipitation strengthened alloys 3 through 7 defined in Table 1 as a result of exposure to a temperature of 200°C. Note that the scale of the ordinate axis is different from that of Fig. 1. shows diagrams summing up Transmission Electron Microscope characterisation of precipitates in alloys 2, 3, 4 and 7 after exposure of a temperature of 200°C for 7 days. Figure 4 a) shows the precipitate number density and volume fraction, whereas Figure 4 b) shows the average precipitate length and cross-section as a function of the effective Mg/Si ratio of the alloys. shows Transmission Electron Microscopy images showing the evolution in microstructrue during high-temperature exposure for alloy 3 and 6; a) and b) shows alloy 3 and 6, respectively, in the T6 condition, c) shows alloy 3 after 2h exposure at 200°C, and d) and e) shows alloy 3 and 6, respectively, after 3 weeks exposure at 200°C. Examples of precipitate types and orientations are indicated on the images, as well as the crystallographic [100] and [010] orientation in the aluminium lattice. depicts a diagram showing the development in strength of precipitation strengthened alloys 3, 8 and 9 as defined in Table 1 as a result of exposure to a temperature of 200°C. Note that the scale of the ordinate axis is different from that of Fig. 1 and Fig. 4. Fig. 7 shows diagrams summing up Transmission Electron Microscope characterisation of precipitates in alloys 3, 8 and 9 after exposure of a temperature of 200°C for 7 days. Figure 7 a) shows the precipitate number density and volume fraction, whereas Figure 7 b) shows the average precipitate length and cross-section as a function of the total content of Mg + Si + Cu in the alloys

Fig. 8 is a diagram showing the development in strength of precipitation strengthened alloys 6 and 10 as defined in Table 1 as a result of exposure to a temperature of 200°C.

Fig. 9 is a diagram showing the development in strength of precipitation strengthened alloys 6, 11 , 12 and 13 as defined in Table 1 as a result of exposure to a temperature of 200°C.

Fig. 10 is a diagram comparing the development in strength of precipitation strengthened alloys 10 and 14 as defined in Table 1 as a result of exposure to a temperature of 200°C. Fig. 11 is a Transmission Electron Microscopy image of alloy 12 after exposure of a

temperature of 200°C for 7 days.

Fig. 12 is a diagram comparing the development in strength of precipitation strengthened alloys 3, 6, 15 and 16 as defined in Table 1 as a result of exposure to a temperature of 250°C.

Fig. 13 is a Mg-Si chart illustrating the Mg and Si contents of claims 1 through 6.

Al-Mg-Si alloys gain their strength from precipitation hardening (also commonly denoted artificial age hardening), a heat treatment by which a fine dispersion of precipitates is formed. The precipitates strengthen the alloy by impeding dislocation movements.

The common temperatures used for precipitation hardening of 6xxx alloys for structural applications lie in the range 150-190°C. As a function of artificial age hardening time the hardness of the alloy will increase to a maximum level and thereafter decrease. When such a decrease in hardness has taken place the condition of the alloy is referred to as "overaged". Exposure of the alloy to temperatues higher than the normal age hardening temperatures gives an acceleration of the mechanisms that lead to overageing

Depending on the alloy composition and the thermo-mechanical treatment applied, several different precipitate types may form. Great advances in the description of the various precipitate types have been achieved over the last decade, which enables a more fundamental approach for alloy development, such as to optimise the alloy for specific properties by formation of specific precipitate types. The common textbook-description of the precipitation sequence in Al-Mg-Si alloys is

SSSS→ GP-zones→ β"→ β'→ β where SSSS denotes supersaturated solid solution, GP-zones is a precursor to the hardening particles, β" is the principal hardening particle, β' is a particle type formed in alloys artificially aged to past their maximum hardness (overaged), and β is the equilibrium phase Mg₂Si. Recent research has shown, however, that a more correct precipitation sequence has the form of

SSSS→ GP-zones→ β"→ β', B' Ul, U2→ β i.e. that there is a range of particle types that may form in the overaged state. Which particle types within β' B' U1 or U2 that form in the overaged state depends on alloy composition and the thermo-mechanical treatment. More often than not, one finds that two or more particle types coexist, but if one of the types is found in significantly higher number density than the others this type is thus the "dominating" particle type for that particular alloy in that particular condition.

It has also been shown that for some combinations of alloy composition and thermo-mechanical treatment the particle types normally associated with an overaged state actually can be dominating in the peak strength condition. So, the sequence of precipitation as indicated above does not take into account all different cases that may arise, but rather gives a general idea of the development of precipitation as a function of temperature and time. Up until recently, it has been the common understanding that small additions of Cu to Al-Mg-Si alloys enhance the strength of the alloy by solid solution strengthening. It has recently been proven, though, that small additions of Cu significantly alters the precipitation sequence to SSSS→ GP-zones→ β", L, S, C→ Q'→ Q where L, S and C are Cu-containing precipitate types that are predominant when the alloy is precipitation hardened to maximum hardness, Q' is a precipitate type characteristic for overaged alloys, and Q is the equilibrium phase in the Al-Mg-Si-Cu system. Which particle type within β", L, S or C that is formed during precipitation hardening depends on alloy chemistry and thermo- mechanical treatment. Just as for the Cu-free alloys, one can find cases where precipitation differs from the sequence above, such as for example that the L-phase can be found to in highest number density, i.e. it is "dominating", in the overaged state for some alloy compositions and thermomechanical treatments.

As stated above, a commonly known aluminium alloy for structural applications in the European market is the standard 6082 alloy. This alloy contains sufficient amounts of Mg and Si for obtaining tensile strength of typically 300-320 MPa under common industrial practices. In addition, the alloy contains a substantial amount of Mn. The Mn forms dispersoids during homogenisation of the alloy. The purpose of the dispersoids is to control the microstructure during thermo-mechanical processing, such as for example obtaining a fibrous grain structure after extrusion of the alloy. The 6082 alloys do, however, lose strength after prolonged temperature exposure at for instance 200°C. In order to find an alloy composition more suitable for high temperature exposure several alloy compositions were investigated, as explained in the examples below. For the purpose of grain structure control, a Mn-level typical for 6082-alloys was chosen for all alloy compositions.

Zr and Cr serve the same purpose as Mn in this type of alloys and could have been used as partial or complete replacements of Mn. For all the examples the alloys were cast as 095mm logs (extrusion ingots), homogenised at a temperature in the range 520 - 600°C for a length of time between 1 and 10 hours, and pre-cut to extrusion billets of 200mm length. The extrusion billets were preheated in an induction furnace, and extruded to cylindrical rods of 20mm diameter. Samples of the extruded rods were solution heat treated in the range 520 - 600°C, water-quenched, stored 4 hour at room- temperature, and then precipitation hardened at 155°C for 12h, corresponding approximately to maximum hardness of the alloy. This is referred to as the T6 condition and this was the starting condition for all the high-temperature exposure measurements in the examples below.

Table 1 - Compositions [wt. %] and effective atomic Mg/Si ratios of the alloys used in the examples of the present application:

effective atomic percent Mg/(Si+Ge)

Example 1.

Fig. 1 shows the development in Vickers hardness as a function of exposure time at 200°C for an alloy of composition 1 in Table 1 above. This alloy represents a typical composition for a 6082-alloy. Within a few hours, the hardness declines rapidly, and after a day the hardness is down to approx. 2/3 of that of the starting condition. Further exposure at 200°C reduces the hardness to 50% of that the starting condition after 1 week.

Simply adding Cu to an alloy of the same composition drastically changes the response to high temperature exposure. Alloy 2 is practically similar to alloy 1 , but with an addition of 0.45 wt.% Cu. As shown in Figure 1 , this leads to a much slower decline in hardness after temperature exposure at 200°C. After 1 week exposure, the decline in hardness is almost reduced to the half as compared to Alloy 1. The reason for this is found in the microstructure of the alloys. Fig. 2 a and b shows transmission electron micrographs of representative precipitate structures of alloys

1 and 2 after exposure at 200°C for 1 week. The precipitates of alloy 1 are much coarser and less densely distributed than those of alloy 2. This explains the lower hardness of alloy 1 as compared to alloy 2. Further, the types of precipitates are different in the two alloys. In the Cu- free alloy 1 one finds a combination of Β', β^', U2 and U1 precipitates. In the Cu-containing alloy

2 the dominating precipitate type is the Q'-phase. It seems that Q' is more stable against precipitate coarsening than Β', β^', U2 and U1.

In the Cu-containing alloys, there are several precipitate types that may be dominating when the alloys are aged to around maximum hardness .. By altering the chemical composition, one may manipulate which precipitate type that is dominating. Both of the alloys 1 and 2 have an effective atomic Mg/Si ratio of 0.84 (The term "effective atomic Mg/Si ratio" refers to the amount of Mg and Si in solid solution prior to precipitation hardening). By keeping the sum (in atomic %) of Mg and Si constant and altering the effective Mg/Si ratio to 2 one gets alloy 3 as shown in Table 1. From Fig. 1 one finds that the hardness of this alloy declines even slower than that of alloy 2 during high temperature exposure. TEM investigations reveal that this is due to an even smaller precipitate size than in alloy 2. Fig. 2 c shows a TEM image of equal magnification as for alloys 1 and 2, after exposure at 200°C for 7 days. Whereas Q' is the dominating precipitate type in alloy 2, alloy 3 contains mostly L-type precipitates. These precipitate types are even more stable against coarsening than the Q'-type, and the features of these precipitates (crystal structure and orientation relationship with the aluminium crystal lattice) are preferred for alloys designed for high temperature stability.

Example 2

In order to find the optimal Mg/Si ratio for high temperature stability a series of alloys, labelled 3 to 7 in Table 1 , with constant atomic percentage Mg + Si but with different Mg/Si ratios was prepared. The results of the development in Vickers hardness as a function of exposure time at 200°C for these alloys are shown in Fig. 3. Alloy 4 with an Mg/Si ratio of 1.5 has the poorest temperature stability. Alloy 7 with an Mg/Si ratio of 4 has the lowest strength in the precipitation hardened state. The thermal stability up to 4 day exposure at 200°C is very good for alloy 7, but after this point the hardness decreases more rapidly than for the other alloys. The highest hardness after long-time exposure at 200°C is achieved for alloys 5 and 6 with Mg/Si ratios of 2.5 and 3, respectively. It thus seems that the preferred alloy composition should be chosen such that the effective atomic Mg/Si ratio lies between 1.5 and 4.

The results of a thorough quantitative TEM characterisation of alloys 2, 3, 4 and 7, all being essentially similar with the exception of the effective Mg/Si ratio, after 1 week exposure at 200°C are summarised in Figure 4. The various precipitate types are identified by diffraction patterns and by characteristic features as observed in the TEM. It is found that the dominating precipitate type in alloy 4 and 7 is the L-type, just as in alloy 3. The number density of precipitates increases with increasing Mg/Si ratio of the alloys. However, the average precipitate length and precipitate cross section is reduced with increasing Mg/Si ratio. For alloy 2, 3 and 4 the volume fraction of precipitates is fairly even, whereas for alloy 7 the volume fraction is significantly lower. In general, precipitation hardening increases with increasing number density and increasing volume fraction of precipitates. The present data on strength evolution after long-time exposure at 200°C and appertaining precipitate statistics clearly show that there is a maximum in high temperature stability for a Mg/Si ratio somewhere between 1.5 and 4 for the investigated alloys.

It is important to point out that the precipitate structure in the T6 condition does not necessarily reflect the precipitate structures that evolves during high temperature exposure. This is shown in Figure 5, where Figs 5 a and b shows the precipitate microstructure in alloy 3 and 6, respectively, in the T6 condition. The crystallographic [100] and [010] orientations of the aluminium lattice are outlined on the pictures. One finds that the precipitate type that dominates are needle-shaped, most of them are of the β" type or precursors to the β" type in both alloys. Some precipitate are indicated with arrows, and some of the needles are indicated to run parallel to the viewing direction (notation "Needle,") and perpendicular to the viewing direction (notation "Needle_± ") A few precipitates of the L-type may also be seen in both samples, but they are not dominating. The orientation of the L-particles is notated in a similar manner as the needles. The microstructures of alloy 3 after 2h exposure at 200°C is shown in Figure 5 c. Note that the magnification is somewhat lower than Figs. 5 a and b. The needle-type precipitates are still dominating in this condition. However, as overageing proceeds the L-particles become more dominant as indicated in Figure 2 c, and after keeping the samples at 200°C for 3 weeks one gets the microstructrue shown in Figure 5 d and e for alloy 3 and 6, respectively. Here, the L- phase is the dominating precipitate type. Looking at the cross-section of the L-particles that have a direction parallel to the viewing direction, one sees a feature that is characteristic for this partcle type: a flat rectangular cross-section with a well-defined long edge aligned in the [100] direction of the aluminium lattice. This is the precipitate type that is most resistant to coarsening, and the key to successfully make a high-temperature stable alloy is that this precipitate type becomes dominant upon high-temperature exposure.

Example 3

In order to find data on the sensitivity of the total content of alloying elements on the

temperature stability of the alloys, two more alloys with effective atomic Mg/Si ratio of 2 were prepared; alloy 8 which has approximately 10% higher content of Mg+Si+Cu than alloy 3, and alloy 9 which has approximately 10% lower content of Mg+Si+Cu than alloy 3, see Table 1 for full composition. Fig. 6 shows the results of the development in Vickers hardness as a function of exposure time at 200°C for these alloys. One finds that the hardness of alloy 3 mostly lies in between those of alloy 8 an 9. For long exposure times the hardness of alloy 3 and 8 approach the same level, whereas the hardness of alloy 9 is clearly lower. This indicates that the sum of alloying elements Mg+Si+Cu chosen for alloy 3 is fairly optimal.

Figure 7 shows the precipitate statistics as measured in TEM for alloy 3, 8 and 9 after 1 week exposure at 200°C. The L-type precipitate was dominating for all alloys in this condition. Alloy 8, with the highest Mg+Si+Cu content, has the highest number densitiy of precipitates, but also the smallest precipitate size, of the alloys. These two differences seem to nearly cancel out on the effect of strength compared to alloy 3 Alloy 9, with the lowest Mg+Si+Cu content has slightly lower number density of precipitates, slightly lower volume fraction, and slightly lower precipitate size than alloy 3. These differences all have a negative effect on strength, and in sum they give a significantly lower strength in alloy 9 than in alloy 3. Example 4

Cu is a critical element in the sense that a high Cu content may be detrimental for processing and fabrication characteristics, such as castability and extrudability of the alloys, as well as for corrosion properties of components. On the other hand, a certain Cu content is necessary to achieve the desired precipitate types in the alloys, namely the precipitate types that have been proven to be most resistant to coarsening during high temperature exposure. In order to find out whether a lower Cu content than 0.4 wt.% is feasible alloy 10 was prepared, which is similar to alloy 6 (Mg/Si ratio is 3) except for the Cu content, which is 0.30 wt.%. The development in Vickers hardness as a function of exposure time at 200°C for alloy 6 and alloy 10 is shown in Fig. 8. The results show that the lower Cu content has a significant negative effect on the strength, and possibly also a slight negative effect on the softening rate of the alloy. The recommended Cu level of the alloy will therefore be dictated by the strength requirement for the desired application.

Example 5

Further advances in thermal stability may be achieved by adding the elements Ag and Ge to the alloy. It has been found that these element may modify the characteristics of the precipitates formed in the alloy, and may further stabilize the precipitates against coarsening during high- temperature exposure of the alloys. Three alloys based on the composition of alloy 6, but with the small addition of 0.12 wt.% Ag (alloy 11 ), substitution of 0.05 wt.% Si with 0.13 wt.% Ge (alloy 12) and with both addition of 0.12 wt.% Ag and substitution of 0.05 wt.% Si with 0.13 wt.% Ge (alloy 13), were prepared. In addition an alloy similar to alloy 10 was prepared, but where a substitution of 0.05 wt.% Si with 0.13 wt.% Ge was made (alloy 14)

Fig. 9 compares the development in Vickers hardness as a function of exposure time at 200°C of the alloys 11 , 12 and 13 to that of alloy 6. It is seen that the substitution of a fraction of Si with Ge (alloy 6 vs alloy 12) enhances the temperature stability of the alloy somewhat. Just adding Ag to the alloy does lower the temperature stability (alloy 11 vs alloy 6), but when used in combination with the Ge it seems to further enhance the temperature stability of the alloy (alloy 13 vs alloy 12)

Fig. 10 compares the development in Vickers hardness as a function of exposure time at 200°C of alloy 14 to that of alloy 10. The substitution of a fraction of Si with Ge leads to a considerable increase in the hardness, and by comparing Fig. 10 with Fig. 8 one finds that the Ge/Si substitution compensates for the lower level of Cu in alloy 10 compared to alloy 6.

TEM investigations indicate that the L-type precipitate is still dominant for the Ge/Ag-modified variants. Figure 11 shows a TEM image of alloy 12 after exposure to 200°C for 1 week. The condition of the material and magnification of the image is identic to those of Figure 2, and can be compared directly. It is evident that the Ge-modification leads to a finer precipitate structure. One recognizes the rectangular cross-sections of the L-precipitates, with the long edge aligned in the crystallographic [100]-direction of the aluminium lattice. Example 6

Testing of high temperature stability was also done at higher temperatures, and findings correlate well with the examples 1-5 above. Figure 12 shows the development in hardness of alloy 3, 6, 15 and 16 during high-temperature exposure at 250°C. Alloys 3 and 6, which lie within the optimal alloy window revealed in this application, have a much better temperature stability than alloys 15 and 16, which are outside this optimal window. It is worthwhile to note that the temperature stability of alloys 3 and 6 at 250°C is much better than that of alloy 2 at 200°C (compare Figs. 1 and 12). This illustrates the great advantage in temperature stability of alloys according to the present invention in comparison with the common 6082-alloy.

TEM investigations of Alloys 3 and 6 after 100 h exposure at 250°C show that the L-phase is the predominant precipitate type. This is the reason why these alloys retain a higher hardness after the high temperature exposure.

Alloys 3 and 6 were also subjected to a high temperature exposure of 350°C for 5h. After this heat treatment there has been a change in dominating precipitate type from L to Q'.

Fig. 13 shows an Mg-Si chart illustrating polygons defined by the coordinates in the Mg-Si diagram visualizing as rectangles the Mg and Si contents as defined in claims 1 through 6.

Claims

An Al-Mg-Si-Cu alloy with good thermal stability, characterised in that its content of Mg and Si lies within a polygon defined by the following coordinates of an Mg-Si diagram:

a1 - a2 - a3 - a4 -a1

where in wt.% a1 = 0.60Mg ,0.60Si, a2 = 0.90Mg, 0.90Si, a3 = 1.30 Mg, 0.60 Si and a4 = LOOMg, 0.30Si,

and with the additional alloying elements:

- Cu between 0.20 and 0.50 wt.%

- Fe between 0.08 and 0.40 wt.%,

and where at least one of the following elements are added for the purpose of grain structure control during processing of the alloy

- Mn between 0 and 0.80 wt.%

- Cr between 0 and 0.30 wt.%

- Zr between 0 and 0.30 wt.%,

and optionally Ti up to 0.1 wt% and B up to 0.1 wt% as grain refining elements, and further optionally Ge between 0 and 0.20 wt.% and Ag between 0 and 0.20 wt.%, rest Al including incidental impurities.

Al-Mg-Si-Cu alloy according to claiml

character izedin that the L-phase is the dominant precipitate type with regard to number density upon over-ageing.

Al-Mg-Si-Cu alloy according to claimsl and 2,

characterised in that

-Mn is between 0.15 and 0.80 wt.%

Al-Mg-Si-Cu alloy according to claim land 2,

characterised in that:

-Mn is between 0.40 and 0.80 wt.% Al-Mg-Si-Cu alloy according to claim 1-4,

characterised in that its content of Mg and Si lies within a polygon defined by the following coordinates in the Mg-Si diagram:

b1 - b2 - b3 - b4 - b1

where in wt.% b1 = 0.70Mg ,0.60Si, b2 = 0.90Mg, 0.80Si, b3 = 1.20 Mg, 0.60 Si and b4 = 0.98Mg, 0.37Si

Al-Mg-Si-Cu alloy according to claim 1 -4,

c1 - c2 - c3 - c4 - c1

where in wt.% d = 0.75Mg ,0.62Si, c2 = 0.90Mg, 0.75Si, c3 = 1.15 Mg, 0.58 Si and c4 = 0.98Mg, 0.42Si.

Al-Mg-Si-Cu alloy according to claims 1 -6

characterised in that

Ge is between 0.004 and 0.20 wt.%.

Al-Mg-Si-Cu alloy according to claims 1 -7

characterised in that

the Mg/Si ratio is between 1.5 and 4.