WO2004001085A1

WO2004001085A1 - Methodof producing formable aluminium alloys

Info

Publication number: WO2004001085A1
Application number: PCT/GB2003/002647
Authority: WO
Inventors: Gary John Mahon; Paul Wycliffe
Original assignee: Alcan International Ltd Canada
Current assignee: Rio Tinto Alcan International Ltd
Priority date: 2002-06-19
Filing date: 2003-06-18
Publication date: 2003-12-31
Anticipated expiration: 2004-12-19
Also published as: AU2003240127A1

Abstract

This invention concerns high strength aluminium/magnesium alloys which can be formed from sheet into the body and ends of beverage containers or into architectural or decorative products such as the panels of automobiles. The production method comprises the steps of a) casting an alloy of a composition comprising the following expressed in weight percent: Magnesium 1.6 to 3.9 Copper 0.1 to 0.6 Manganese 0.05 to 1.2 Iron up to 0.8 Silicon up to 0.4 Chromium up to 0.2 Titanium up to 0.2 Boron up to 0.05 Zinc up to 0.2 Balance: Aluminium with incidental impurities, b) optionally homogenising the cast alloy, c) hot working the casting at an initial temperature of at least 400 °C to form an intermediate product, wherein at least part of the hot working is carried out whilst the casting is at a temperature above the solvus temperature of the alloy, d) cooling the intermediate product either during hot working or in a subsequent step at a rate such that at least a proportion of the dissolved magnesium and copper in the intermediate product is precipitated out on a substantially recrystallisation-free grain structure in the form of a substantially uniform dispersion of particles or clusters of particles of a size sufficiently small that during the alloy's subsequent thermal history the alloy undergoes substantial dispersion (precipitation) hardening, and e) optionally annealing the cooled intermediate product to form therein a partially annealed structure.

Description

METHOD OF PRODUCING FORMABLE ALUMINIUM ALLOYS

This invention concerns high strength aluminium/magnesium alloys which can be formed from sheet into the body and ends of beverage containers or into architectural or decorative products such as the panels of automobiles .

Current sheet used for the manufacture of beverage containers, such as cans, uses alloys such as AA3004 or AA3104, which obtain their strength in large part due to strain hardening during cold rolling. This strength can increase further during working of the sheet during container manufacture. Since these materials are highly strain hardened they have low formability, as can be seen from parameters such as tensile ductility or Erichsen dome height. In order to improve formability the sheet or container can be annealed. This can either be a recovery anneal, which only increases formability slightly, or a recrystallisation anneal, which will produce a much more formable material . The downside of the recrystallisation anneal is that the strength drops to a fraction of the initial strength of the sheet.

Alternatively a partial anneal can be performed to give a better balance of strength and formability, but in practice this can be difficult to achieve due to the effects of time and temperature on the extent of recrystallisation and due to the variability of conditions in the sheet manufacturing process.

If a full or partial anneal were to be used during manufacture of the containers made from AA3004, AA3104 or AA5182 type alloys, the strength would drop to such a large extent that the material would need to be of a much thicker gauge, thereby increasing the cost of the containers.

The present invention seeks to provide a method for manufacturing a sheet suitable for the production of beverage containers (such as cans) using conventional drawings and wall ironing technology, or for the manufacture of beverage can ends. In this invention, the sheet is strengthened by a combination of precipitation strengthening, work hardening and solid solution strengthening without the need for a solution heat treatment step with a fast quench.

In US-A-5913989 a process for manufacturing aluminium can body stock is described using an aluminium alloy containing magnesium in the range 1.1 to 1.5% by weight. When referring to the magnesium content, it is stated that if it is too high the final product will undergo excessive work hardening during drawing and ironing and be more prone to scoring. It has now been found that there is advantage is exceeding that 1.5% limit so as to obtain the benefits of precipitation hardening.

US-A-5822843 and 6010028 both describe using hard rolled thin gauge sheet as the starting material for producing beverage cans. By contrast the present invention does not utilise such hard rolled sheet.

JP60-50141 (S58-156758) describes the formation of can ends from an aluminium alloy hard sheet containing 2.0 to 5.0% magnesium and 0.05 to 0.5 weight percent copper. It is, however, described in that specification as essential to use an interanneal in order to keep the magnesium and copper in solid solution. By contrast, in the present invention precipitation of magnesium and copper is desired.

JP10-121179 (H8-279462) describes a similar process to that of JP60-50141, except that the magnesium content is from 2.5 to 3.5% and the copper content is from 0.1 to 0.5%. Again, however, that invention requires that the magnesium and copper are maintained in solid solution during the main cooling step.

In contrast to the known prior art, during manufacture of sheet in accordance with the present invention the processing conditions during hot working and cooling are controlled such that a dispersion of fine precipitates is formed at that stage . These precipitates may take the form of particulate clusters which form in the deformed microstructure on dislocations or subgrain boundaries . Such clusters may not be readily observable using transmission electron microscopy, but their presence can be detected because the precipitates give a high strength to the sheet when it is cold rolled to final gauge.

This precipitation of fine particles is not the same as the precipitation of S phase or its metastable precursors which is described in W0 02/50329. In that prior process the Cu in the alloy must be retained in solid solution during the cooling step so as to provide an age hardening effect. There is no recognition in WO 02/50329 of the possible use of precipitation hardening as a substantial strengthening mechanism. Any particulate precipitation and resultant hardening effect which may occur if this prior method is adopted will in practice be insignificant . Similarly another can stock manufacturing method is described in WO 99/39019 in which a minor amount of particulate precipitation may occur as an unintended consequence of the cooling step used. In that prior method improved strength is said to be achieved by solid solution and age hardening, and for that purpose a quench step is described as essential in order to avoid substantial particulate precipitation. Particulate precipitation is mentioned in WO 99/39019 but only in the context of a batch stabilisation step after rolling to final gauge. There is no recognition in WO99/34019 of using particulate precipitation as a substantial source of hardening.

The method of the present invention and its preferred aspects are set out in the accompanying claims. Analysis of strengths obtained by the present invention compared with the strength obtained in the absence of precipitation shows that strength increases of approximately 30 Mpa can be achieved in addition to the strength provided simply by work hardening plus solid solution strengthening. This strength is achieved without the requirement of a costly rapid quench step in the processing.

For sheet for can end stock, a final gauge yield strength of at least 370 MPa is desired. Currently for can end stock made of aluminium alloy AA5182, the strength is achieved by a combination of work hardening (90% cold rolling reduction to final gauge) and solid solution strengthening (through the use of approximately 4.5% by weight Mg in the alloy) with no substantial contribution from precipitation strengthening. The high Mg content increases the cost of the AA5182 alloy and makes hot rolling more difficult since the solid solution strengthening due to Mg increases rolling loads. In the present invention the alloy is less costly, the loads encountered during hot rolling are reduced in comparison with AA5182, and the contribution by precipitation strengthening gives yield strengths in excess of 400 MPa (despite the alloy' s lower level of Mg) .

For sheet for can end stock, the material of the present invention, preferably in sheet form, may be used in a fully hard temper, e.g. H19, where the material has high strength due to a combination of precipitation strengthening, solid solution strengthening and work hardening. Alternatively, the material can be cold rolled to a final gauge and given a partial anneal or a stabilization anneal to give a balance of high strength with increased formability.

In order to achieve the desired strength and microstructure in the starting sheet, its composition should be in the following ranges expressed in weight percent .

Magnesium 1.6 to 3.9

Copper 0.1 to 0.6

Manganese 0.05 to 1.2

Iron up to 0.8

Silicon up to 0.4

Chromium up to 0.2

Titanium up to 0.2

Boron up to 0.05

Zinc up to 0.2

Balance Aluminium with incidental impurities, The alloy is cast and processed so that substantially all of the Mg and Cu are in solid solution prior to the final stages of hot rolling, or prior to extrusion if that production method is used. The processing steps include optionally homogenising the cast alloy, hot working the casting at an initial temperature of at least 400°C to form an intermediate product, wherein at least part of the hot working is carried out whilst the casting is at a temperature above the solvus temperature of the alloy, and cooling the intermediate product, either during hot working or in a subsequent step, with all the final stages of deformation being carried out at a temperature below the recrystallisation temperature for that alloy.

During cooling the intermediate product is exposed to a temperature preferably in the range 270 to 180° C so that at least a proportion of the dissolved Mg and Cu in the intermediate product can precipitate out from solid solution on a substantially recystallisation-free grain structure (i.e. worked) such that a substantially uniform dispersion of precipitates is achieved through the aluminium matrix. Optionally or additionally the last stages of hot working (e.g. the final rolling passes) may be at lower temperatures (e.g. 100 to 320°C, preferably 180 to 270°C) in order to provide a fine scale recovered but unrecrystallized deformation substructure in the intermediate product that facilitates particulate precipitation. Reduced temperatures of working promote a fine scale substructure which in turn yield a finer scale of precipitate which leads to an increase in strength. The average particle size of the precipitates should be sufficiently small, i.e. less than 0.2 microns, that during the alloy' s subsequent thermal history the alloy undergoes substantial dispersion (precipitation) hardening. Preferably the average size should be less than 0.07 microns and more preferably less than 0.05 microns . The intermediate product can then be further rolled to the final gauge to provide a fully hard product. Optionally the sheet can be annealed at final gauge to provide either a partially annealed sheet or stabilised sheet. Any such low temperature anneal should be performed under conditions substantially to leave in the matrix a fine dispersion of precipitates capable of providing hardening.

Although the hot working step could be carried out by extrusion, generally rolling is used to form a sheet.

The objective of the controlled cooling step, whether it is cooling at the re-roll stage or a partial inter-anneal during cold rolling, is to maintain the peak temperature of the material low enough so that the alloy remains in an unrecrystallized state. This promotes the development of a fine precipitate structure during the slow cooling phase, which in turn leads to an increased strengthening effect. It is the combination of the low temperature of the final stages of hot working, low coiling temperatures, (low peak anneal temperatures for interannealed material) and slow cooling within the specified alloy range that gives the optimized strengthening effect, not predicted by the prior art. Preferably the alloy has the following composition expressed in weight percent :

Magnesium 2.0 to 3.8

Copper 0.25 to 0.55

Manganese 0.1 to 0.5

Iron up to 0.4

Silicon up to 0.2

Chromium up to 0.1

Titanium up to 0.2

Boron up to 0.05

Zinc Up to 0.2

Balance Aluminium with incidental impurities

A particularly preferred alloy composition in weight percent is:

Magnesium 2.85 to 3.15, preferably 3.00

Copper 0.35 to 0.45, preferably 0.40

Manganese 0.20 to 0.30, preferably 0.25

Iron 0.15 to 0.25, preferably 0.20

Silicon 0.08 to 0.16, preferably 0.12

Balance Aluminium with incidental impurities

Preferably the total of the incidental impurities is less than 0.5% by weight.

Desirably most of the hot working is carried out at a temperature above the solvus temperature of the alloy and hot reversing rolling is used. The final passes of hot rolling are below the recrystallization temperature of the alloy, and may take place on a hot tandem mill. Preferably the temperature of the sheet exiting the hot rolling is between 100 and 320°C, and preferably between 180 and 270°C. Desirably the cooling rate for the exiting hot sheet is between 2 and 400°C/hr., and preferably from 0.1 to 1.5°C/min.

The sheet may be used in a hard, e.g. H19, temper for beverage can ends. Formability can be improved by using a partially annealed temper without sacrificing strength. Examples of different tempers which can be used are:

1) Fully hard which means as-cold rolled (i.e. H19) ,

2) Partially annealed by cold rolling to final gauge and then softening by adjusting the sheet temperature after cold rolling to about 130 to 160°C, or

3) Partially annealed by a formal partial anneal in a furnace, such as holding for 1 hour or longer at 120 to 240°C.

The present invention will now be described in more detail with reference to the following illustrative example taken in conjunction with the accompanying drawings in which:- Figure 1 is a graph showing the variation of electrical conductivity with time for reroll for different reroll anneals,

Figure 2 is a graph showing the variation of reroll yield strength with time for different reroll anneals,

Figure 3 is a graph showing the relationship between final gauge yield strength and reroll yield strength,

Figure 4 is a graph showing the variation of final gauge yield strength with time for different reroll anneals,

Figure 5 is a graph showing the variation of final gauge yield strength with anneal temperature for different reroll anneals,

Figure 6 is a graph showing the variation of temperature with time for a transfer slab during hot rolling,

Figure 7 is a similar graph to Figure 6,

Figure 8 is a graph showing the variation of temperature with time for the coil cool of reroll after hot rolling,

Figure 9 is a similar graph to Figure 8, and

Figure 10 is a graph showing the variation of final gauge yield strength with temperature for isothermal aging and with coiling temperature after hot rolling. An alloy of the following preferred composition was tested:

Table 1

Laboratory processing was carried out on a number of samples to duplicate the sequence of microstructural evolution that would take place in an industrial process according to the present invention, viz.

1. Hot/warm rolling leading to a reroll with the sheet in an unrecrystallized condition,

2. Solid state precipitation in the reroll, and

3. Cold rolling to final gauge.

Two separate studies were performed:

A. An isothermal aging study on solution heat treated "reroll" that contained 50% cold work.

B. Laboratory hot rolling followed by simulation of coil cooling in order to assess the effect of hot rolling finish temperature.

The isothermal aging study is the more controlled experiment that involves precipitation on a cold worked microstructure with a systematic variation of times and temperatures. Recovery of the deformed microstructure and solid state precipitation will occur simultaneously in this experiment while the specimen is being re-heated, and during the isothermal hold. The cold worked microstructure prior to aging will be of a finer scale than that produced in hot rolling.

Laboratory hot rolling followed by simulation of coil cooling is a more direct demonstration of the effect of hot rolling finish temperature than isothermal annealing. Dynamic recovery during warm rolling will lead to a more recovered microstructure than cold rolling. Further, there is no excursion to room temperature prior to the aging in this experiment. The differences from commercial hot rolling is that the reductions per pass are smaller, and the speeds and strain rates are lower.

The deformation structure prior to precipitation is (for a given exit temperature) likely to be coarser for laboratory rolling than for commercial rolling.

Assuming that the degree of recovery of the deformed microstructure influences the scale of the precipitates that form, then for a given simulated exit temperature in comparison with commercial tandem mill rolling, the isothermal aging experiments will give finer precipitates while the laboratory hot rolling will give coarser precipitates. In terms of strength, the isothermal aging is an upper bound estimate while the laboratory hot rolling is a lower bound estimate. Experimental Part A: Isothermal Aging Study

Laboratory processing was performed on a number of samples in the following sequence:

1. A transfer slab (28mm thick) of the alloy was preheated to 520°C for 1 hour and hot rolled to 5 mm. (That is twice reroll gauge) .

2. At 5mm the material was solution heat treated: 1 hour at 520°C then water quenched.

3. The SHT material was then cold rolled to 2.54mm. (50% thickness reduction to "reroll" gauge) .

4. Isothermal Transformation Aging curves were generated at reroll gauge:

Temperatures: 320, 280, 240, 200°C Times: 0.20 min, 3 h, 8 h, 24 h

Measurements performed on the reroll were electrical conductivity and tensile properties.

5. The aged reroll was then cold rolled to final gauge

(0.25mm) and heated for 5 minutes at 120°C before cooling at 24°C/h to 65°C.

Final gauge tensile properties were determined at room temperature . Table II shows final gauge tensile test results for the Isothermal anneal study.

Table II

Final Gauge Yield Strength vs . Reroll Anneal Time and Temperature

Examples of high strengths according to the present invention are marked with an asterisk in Table II. The higher temperatures in Table II, i.e. 280 and 320°C, are more typical of those of coils after hot rolling and are higher than the preferred range of the present invention and give lower final gauge yield strengths .

The variation of electrical conductivity with time is shown in Figure 1. Each conductivity measurement combines effects due to recovery of the cold worked structure with effects due to precipitation. It will be noticed that the conductivity after 24 hours of aging is reduced as temperature decreases, and that the extent of aging and recovery is less at the lower aging temperatures. Figure 2 shows the variation of reroll yield strength with time. The 200°C heat treatment shows evidence of a peak-aged condition. All other temperatures show monotonic softening due to the combination of recovery and overaging. Softening is time dependent as would be expected for overaging and for recovery. This implies that cooling rate is important.

Figure 3 shows the relationship between final gauge yield strength and reroll yield strength. It will be noted that higher reroll strength gives higher final gauge strength.

Figure 4 plots the variation of final gauge yield strength with time. Again, this implies that there is dependence on cooling rate to avoid too much overaging.

Figure 5 plots the relationship between final gauge yield strength and anneal temperature. Although this plot is from isothermal aging treatments, it suggests the effect of coiling temperature on final gauge properties. The steepness of the curve is noteworthy.

Experimental Part B: Hot Rolling Study

Laboratory hot rolling followed by simulation of coil cooling was used to assess the effect of hot rolling finish temperature. These experiments were a more direct demonstration of the effect of hot rolling finish temperature than the isothermal annealing study in part A. Laboratory processing was performed in the following sequence :

1. A transfer slab (28 mm thick) of the alloy was preheated to 520°C for 1 hour, and hot rolled to

2.5 mm in 4 passes.

2. Hot rolled strip (reroll) was immediately transferred to a furnace where it was given a coil cooling simulation.

3. The reroll was then cold rolled to final gauge

(0.25 mm) and heated for 5 minutes at 120°C before cooling at 24° C/h to 65°C.

Two different variants were produced which had different hot rolling finish temperatures along with different start temperatures for the coil cooling simulation. These are shown in Table III below with detailed temperature vs. time data shown in Figures 6 to 9.

Figures 6 and 7 show the temperature profile for the transfer slabs of Tests 4 and 6, respectively, both of which were subjected to hot rolling on a laboratory mill using a four pass schedule. For both slabs the initial gauge was 28mm, but for Test 4 the final gauge was 2.52mm while for Test 6 the final gauge was 2.73mm. Figures 8 and 9 show coil cool simulations for reroll after hot rolling on a laboratory mill for Tests 4 and 6, respectively. For Test 4 the coil cool was from 280°C for the 2.52mm sample, whilst for Test 6 the coil cool was from 240°C for the 2.73mm sample. The mill exit temperature for Test 4 was 320°C whilst that for Test 6 was 301°C.

Precise control of hot rolling finish temperatures is difficult in laboratory hot rolling, and the decrease in temperature from the reroll exit temperature to the start of the coil cooling is much larger than for commercial rolling.

Table IV shows final gauge mechanical properties for the material after it was cold rolled to final gauge (0.25mm) and given simulation of cooling for cold rolled coils.

Table III

Hot Rolling Temperatures

Table IV

Figure 10 shows the variation of final Gauge Yield Strength with Temperature for both the Isothermal Aging study and the laboratory hot rolling study. Temperatures for the Isothermal aging data (IT) are the aging temperatures, while temperatures for the hot rolling data are the coil cooling simulation start temperatures. It will be noted that both sets of data give similar results; strength increases by about 30 MPa for a decrease in coil temperature from 280 to 240° C. This shows the dependence of final gauge yield strength on hot mill coiling temperature. It will be noted that final gauge strengths are near 400 MPa for cold rolling finish temperatures of 120°C.

Estimate of the strength contribution due to Precipitation hardening

Table V below shows the evolution of strength through the laboratory processing for some of the samples . The observed effect of reroll aging treatment has two alternative explanations :

1. The lower aging temperature (simulating a lower coil temperature) gave a high strength at final gauge because there was less softening of the reroll, and as a result there was more work hardening at final gauge to give a high strength, or

2. There was some precipitation hardening taking place, and the precipitation hardening contributed to the strength at final gauge .

Closer examination of the data shows that the latter explanation is correct. With reference to Table V, the column headed "no age" describes material that was cold rolled from 5 mm to 2.54 mm to 0.25 mm. Cold rolling the material from 5 mm to 2.54mm increases the strength from 94 MPa to 300 MPa.

Further rolling to 0.25 mm increases the strength to 394 MPa. The final gauge strength for this material is a result of work hardening only.

The material labelled "3h at 200°C" used the same starting material and its strength was decreased by aging the 2.54 mm thick reroll for 3 hours at 200°C. This strength reduction occurs because softening due to recovery is a stronger effect than the precipitation strengthening that occurred. However, upon further cold rolling to the final gauge of 0.25 mm, this material increased in strength to 423 MPa. The final gauge strength was higher than that for material that was only work hardened. This indicates the presence of an additional strengthening mechanism, i.e. precipitation strengthening.

The trend in final gauge yield strength may be observed for other aging treatments shown in Table V. Upon cold rolling to final gauge, the material given the aging treatment is stronger than material that is simply cold rolled, despite the fact that at the reroll gauge the aging had softened the material .

Comparison of the final gauge yield strengths in Table V gives a quantitative estimate of the contribution made by precipitation strengthening. The sample that had "no age" had a yield strength of 394 MPa while samples aged for 0.33 h and 3 h at 200°C had yield strengths of 426 MPa and 423 MPa, respectively. Assuming the level of strength due to work hardening was the same for all three samples, the contribution made by precipitation strengthening could therefore be assessed as 32 and 29 MPa, respectively, for these two samples.

TABLE V Evolution of Yield Strength during Processing

Examination by Transmission Election Microscopy

Transmission electron microscopy was used to examine particle sizes in a number of samples at 2.54 mm gauge from the processing sequence described above . The TEM samples were cold rolled from 5 to 2.54 mm and then aged for 3 hours at 200, 240 or 320°C. Precipitation occurs on the recovered deformation substructure so that precipitates occur on the walls of dislocations that form subgrain boundaries .

The sample aged at 320°C was recrystallised and contained particles from 0.1 to 0.3 microns in size. The recrystallisation had removed all subgrains and there was no relation between any existing subgrain structure and the precipitate particles. Particles this large do not contribute as much to strength as finer particles.

A temperature of 320°C is a typical temperature for a coil of metal at the exit of a hot rolling mill that leads to fully recrystallized hot rolled metal, sometimes referred to as "self-annealed" . The use of this higher temperature represents a comparative example and illustrates that high strengths are not obtained if coiling temperatures are used which are above the range required by the present invention.

The sample aged for 3 hours at 240°C was not recrystallised and had a structure consisting of subgrains about 1 micron in size and also included particles from 0.04 to 0.07 microns in size.

For the lowest aging temperature of the three samples particles were not visible because of the high dislocation density which dominated the contrast in any images taken. Extrapolation of the observations on the samples from 320° and 240°C suggests that finer particles were present at the lowest of the three temperatures .

Claims

CLAIMS 37351.GBA - Specifications

1. A method of producing a formable aluminium alloy comprising the steps of: -

a) casting an alloy of a composition comprising the following expressed in weight percent : -

Magnesium 1.6 to 3.9

Copper 0.1 to 0.6

Manganese 0.05 to 1.2

Iron up to 0.8

Silicon up to 0.4

Chromium up to 0.2

Titanium up to 0.2

Boron up to 0.05

Zinc up to 0.2

Balance Aluminium with incidental impurities, b) optionally homogenising the cast alloy,

c) hot working the casting at an initial temperature of at least 400°C to form an intermediate product, wherein at least part of the hot working is carried out whilst the casting is at a temperature above the solvus temperature of the alloy,

d) cooling the intermediate product either during hot working or in a subsequent step at a rate such that at least a proportion of the dissolved magnesium and copper in the intermediate product is precipitated out on a substantially recrystallization-free grain structure in the form of a substantially uniform dispersion of particles or clusters of particles of a size sufficiently small that during the alloy' s subsequent thermal history the alloy undergoes substantial dispersion (precipitation) hardening, and

e) optionally annealing the cooled intermediate product to form therein a partially annealed structure.

2. A method as claimed in claim 1 wherein the alloy has the following composition expressed in weight percent:

Magnesium 2.0 to 3.8

Copper 0.25 to 0.55

Manganese 0.05 to 0.5

Iron up to 0.4

Silicon up to 0.2

Chromium up to 0.1

Titanium up to 0.2

Boron up to 0.05

Zinc Up to 0.2

Balance Aluminium with incidental impurities,

3. A method as claimed in claim 2 wherein the alloy has the following composition expressed in weight percent:

Magnesium 2.85 to 3.15

Copper 0.35 to 0.45

Manganese 0.20 to 0.30

Iron 0.15 to 0.25

Silicon 0.08 to 0.16

Balance Aluminium with incidental impurities

4. A method as claimed in any one of the preceding claims wherein the intermediate product is in sheet form.

5. A method as claimed in any one of the preceding claims wherein all of the hot working is carried out at a temperature above the solvus temperature of the alloy.

6. A method as claimed in any one of the preceding claims wherein the hot working is effected by rolling, preferably hot reversing rolling.

7. A method as claimed in any one of the preceding claims wherein the intermediate product is coiled after the hot working step.

8. A method as claimed in claim 7 wherein the coiling step is part of the annealing step.

9. A method as claimed in any one of the preceding claims wherein the latter stages of the hot working step (c) are carried out at temperatures in the range 100 to 320°C in order to produce in the intermediate product a fine scale recovered but unrecrystallized deformation substructure that facilitates particulate precipitation.

10. A method as claimed in any one of the preceding claims wherein the temperature of the intermediate product exiting the hot working step is between 100 and 320°C.

11. A method as claimed in claim 10 wherein the exit temperature is between 180 and 270°C.

12. A method as claimed in any one of the preceding claims wherein the cooling rate in step (d) is between 2 and 400°C/hour.

13. A method as claimed in claim 12 wherein the cooling rate is 0.1 to 1.5°C/min.

14. A method as claimed in any one of the preceding claims including the step of forming the intermediate product by cold rolling.

15. A method as claimed in claim 14 wherein the formed alloy is in its H19 temper.

16. A method as claimed in any one of the preceding claims wherein the intermediate product is formed into the body or end of a beverage can.