The present invention relates to a method for the manufacturing of extruded profiles that can be anodized with high gloss surfaces, the profiles being extruded of an age hardenable aluminium alloy that can be recrystallized after cold deformation, for example 6xxx (Al—Mg—Si) alloys or 7xxx (Al—Mg—Zn) alloys.
The oxide layer (Al2O3) formed during anodizing is build up by dissolving the outer layer of the aluminium. For each 3 μm of oxide layer formed 2 μm of the aluminium is dissolved. Since the oxide layer is bulkier than the aluminium the total thickness will then increase by 1 μm. In order to obtain high gloss of an anodized aluminium product it is important to keep the amount of constituent particles with a diameter larger than approximately 0.3 μm (S. Wernick, R. Pinner and P. G. Sheasby, The Surface Treatment and Finishing of Aluminium and its Alloys, ASM INTERNATIONAL, FINISHING PUBLICATIONS LTD, Fifth Edition Vol 1, 1987, p. 143) at a low level, since these particles will be embedded in the anodized layer and cause a reduction in the gloss. The most important factor to achieve this is to keep the amount of Fe at a low level, since primary AlFeSi particles are insoluble in the aluminium matrix. Typically, alloys used for high gloss products have a maximum limit of Fe around 0.12 wt %. Gloss is thus also reduced with increasing thickness of the oxide layer formed during anodizing since more particles then will be embedded. Moreover the process parameters used during anodizing also affect the gloss.
Hardening precipitates are formed during the artificial ageing process (e.g. β″-MgSi) from the addition of Mg and Si. If Cu is added in sufficient amount other phases than β″ may form (e.g. Q′ and L) (Calin D. Marioara, et. al., Improving Thermal Stability in Cu-Containing Al—Mg—Si Alloys by Precipitate Optimization, METALLURGICAL AND MATERIALS TRANSACTIONS A, March 2014). These hardening precipitates are much smaller than 0.3 μm and are therefore not reducing the gloss in the same way as the primary AlFeSi particles. The strength requirement for the alloy determines the necessary amount of Mg, Si and Cu in the alloy. In order to maximize the gloss it is necessary to process the material in a way where precipitation of larger non-hardening phases (e.g. β′-MgSi and β-Mg2Si) of Mg, Si and Cu is avoided. This is easiest to obtain for 6060 and 6063 type of alloys where the Mg and Si contents are relatively low. Higher alloyed material requires higher temperatures in the extrusion or solutionising processes and faster cooling afterwards to avoid precipitation of such particles.
Alloying elements such as Mn, Cr, Zr or Sc can be added to form dispersoid particles during homogenization. Frequently, these elements are added in high amounts in order to prevent recrystallization in the extruded profile. However, it can be beneficial to add these elements in smaller amounts to only have some dispersoid particles in the alloy in order to avoid grain growth during homogenization and after the recrystallization process occurring in the extrusion process or in a separate recrystallization and solutionising process for the cold deformed material. The size of these particles is typically between 0.01-0.2 μm. Thus, such particles can be added, at least in a relative low number, without significantly affecting the gloss. However, the number of dispersoid particles should not be so high that the exposed areas of the profile surface get a mixture of a non-recrystallized and a recrystallized structure or a fully recrystallized structure with a large and uneven grain size. Addition of elements that form dispersoid particles can also give an unwanted color of the anodizing layer, or they can give an unwanted surface appearance due to a strong texture of the recrystallized grains.
If an anodized surface contains large grains the individual grains can be detected by the naked eye. This surface defect is frequently called mottling. The best surface appearance is obtained when the average grain size is smaller than approximately 70 μm and the grains mainly are randomly orientated.
If the processing of the material is satisfactory there will be no large β′-MgSi or β-Mg2Si particles present in the extruded and aged profile samples. In such a case the gloss will be more or less proportional to the amount of Fe in the alloy for a given anodizing process. To maximize the gloss one would like to minimize the Fe content. Reducing the Fe content will increase the price of the aluminium since it will be more costly to produce. It will require alumina with low Fe and low contribution of Fe from the anodes. The processing in the electrolysis and the casthouse also has to be adapted in order to produce aluminium with very low Fe content. The main problem by using very low Fe contents is, however, the ability to control the grain size in the billet and in the extruded profile.
From Japanese patent publication No. 10-306336 is known an aluminium alloy extruded material having high surface gloss after anodic oxidation treatment where the surface gloss allegedly is made uniform by specifying the number of the particles of Mg2Si participated in the matrix. This is obtained with a specific heat treatment procedure prior to and after extrusion.
With the present invention it is provided a method for the manufacturing of products that can be anodized high gloss surfaces from extruded profiles of for example 6xxx or 7xxx alloys, with excellent mechanical properties and at low costs.
The method according to the invention is characterized by the features as defined in the accompanying independent claim 1. Further embodiments are defined in the subordinate claims 2-14.
The invention will be further described in the following by way of examples and with reference to the drawings and figures where:
When the Fe content is below approximately 0.10 wt % in a 6060 or 6063 type of alloy the chance of getting abnormal grains (grains that grow and consume other grains that were formed during casting) in the billet during homogenization becomes very high. Therefore, a grain size of several centimeters is very common in billets of alloys with very low amounts of Fe. An example of abnormal grains in a homogenized billet with low Fe content is shown in
For a 7xxx alloy the homogenizing temperatures are normally lower than for a 6060 alloy. This may reduce the problem with abnormal grain growth during homogenization.
The billet grain size will probably not affect the grain size in the extruded profile much if the extent of deformation is high, for example when extruding thin walled hollow profiles. For solid shapes, and especially for thick walled profiles, the billet grain size will most likely affect the grain size in the extruded profile. An additional challenge is that the billet temperature needs to be rather high in order to dissolve the Mg2Si particles, and a high billet temperature makes it more difficult to obtain a small grain size after extrusion.
In an extruded profile, one usually sees a surface layer of mainly randomly oriented grains and typically one or a few grains in thickness. Underneath this layer one typically finds a region of larger grains. The thickness of this layer varies, and is usually thicker for a thick walled and wide solid shape profile and thicker towards the back end of the extruded length. An example of a typical grain structure in a cross section of a thick walled industrially extruded profile can be seen in
More and more consumer electronics like mobile phones, tablets and lap tops are made of aluminium from extruded profiles. If the profile surface could have been used without any machining the grain structure in the anodized surface would probably be okay in most cases. However, very often there is a need to machine the extruded profile to make the shape and the dimensional tolerances of the final product. In that case the exposed surface can consist of grains from the coarse grain layer beneath the surface layer of the extruded profile. Due to this the entire coarse grain layer has to be removed before starting to machine the shape of the final product. The thickness of the layer that has to be removed due to coarse grains will vary with the width of the profile and the extrusion conditions and is typically in the range of 0.2 to 1 mm.
The present invention deals with the task to get a homogeneous grain structure with an average grain size below approximately 70 μm irrespective of the Fe content, the grain size in the billet prior to extrusion and the extrusion conditions.
Solid shape profiles which are blanks for consumer electronics will be more or less flat, but could possibly have some features in the cross section in order to save material and machining. Such profiles are therefore very well suited for cold rolling after extrusion. By cold rolling a profile by a minimum of 10% followed by flash annealing a new recrystallization process will take place. With sufficient deformation and a proper annealing process the resulting grain structure will be homogeneous over the cross section with a much more random orientation of the grains than in the as extruded profile. The grain size will in addition to the alloy content, depend on the degree of cold deformation, the annealing temperature, the heat up conditions and the time at the annealing temperature. In an alloy with very low Fe and no dispersoid particles the recrystallization will take place at a low temperature, most likely during heating to the annealing temperature. One issue will then be to avoid grain growth at the annealing temperature when there are almost no particles in the material to pin the grains.
The annealing temperature should preferably be above the solvus temperature for Mg2Si particles (6xxx) or Zn2Mg particles (7xxx) in order to avoid particles that can reduce the strength and the gloss of the anodized material. In addition, the time at this annealing temperature should be as short as possible in order to avoid grain growth. Therefore, the material should be processed through extrusion in a way that Mg2Si or Zn2Mg particles are avoided. This means sufficiently high billet temperature in combination with a high enough exit temperature from extrusion and fast cooling of the profile after extrusion. With no Mg2Si or Zn2Mg particles in the material prior to cold rolling and annealing there is no need for a holding time for the material at the annealing temperature.
The consequence of annealing at temperatures below the solvus temperature will be that Mg—Si containing precipitates or Zn2Mg precipitates larger than approximately 0.3 μm may form. These particles will contribute to a reduction in the gloss and in the strength of the material. The amount of this reduction will depend on the actual time-temperature history during the flash annealing and cooling operation and the composition of the alloy.
An industrial process to perform the cold rolling and the annealing process could be done as shown schematically in
In order to avoid precipitation of Mg—Si containing precipitates or Zn2Mg precipitates larger than approximately 0.3 μm the profile needs to be cooled rather rapidly down to room temperature. The reason for this is described in a previous section. Thus, preferably according to the present invention, the profile is flash annealed with a heating time of maximum two minutes to a temperature of between 450-530° C. and held at this temperature for not more than 5 minutes and subsequently quenched.
After the annealing operation there is probably a need to remove residual stresses from the quenching operation. The best way to this would probably be to stretch the material in way similar to what is done after extrusion.
After the annealing process the final ageing of the material can for example be done with the patented dual rate ageing cycle (U. Tundal and O. Reiso, EP 1 155 161 B1) to get maximum strength with minimum amount of alloying elements.
The invention will be further described in the following by way of examples.
Billets with diameter 95 mm were cast in a lab casting facility using the Hycast hot-top gas-slip technology (as described in EP 0 778 097 B1) and a TiB2 based grain refiner. The composition of the alloy is shown in Table 1.
The billets were homogenized at 575° C. for 2 hours and 15 minutes followed by cooling at a rate of approximately 400° C. per hour. Extrusion of the billets was performed at an 8 MN laboratory extrusion press with a 100 mm diameter container to a profile with 5×40 mm2 cross section. The billet preheating temperature was approximately 500° C. and the extrusion speed 20 m/min. After extrusion the profile was quenched in water.
A 50 cm long piece from the front part of the extruded profile was cold rolled to give 10, 20, 40 and 60% reduction in the thickness. The samples that were cold rolled to different thicknesses were then annealed in a salt bath which had been preheated to 500° C. A hole was drilled into each of the samples to fit a thermocouple. The heating time to temperature was in the range 5-10 seconds, depending on the thickness of the sample. When a sample was put into the salt bath a holding time of 10 seconds started when the temperature reached 490° C. After annealing the samples were quenched in water.
The cross section of all samples (in all examples) were prepared by grinding and mechanical polishing with a final step using 1 μm diamond paste. In order to make the grains visible in polarised light, anodising was performed in a Struers Lectropol-5 with the following parameters. Voltage: 45 V; Flow rate: 3; Temperature: −5° C.; Time: 2 minutes. The electrolyte had the following ingredients: 74% distilled water; 24% ethanol; 1% HBF4 (35%); 1% HF (40%).
Prior to extrusion the billets had an even and small grain size. The as extruded sample in
As can be seen from the large color contrast, the cold rolled and annealed samples show a much more random orientation of the grains than the as extruded sample. This confirms that these samples are fully recrystallized after annealing. The samples that were cold rolled to 10 and 20% reduction in thicknesses clearly have an uneven grain structure with the largest grains in the middle of the cross section. The samples that were cold rolled to 40 and 60% reduction in thicknesses have an even grain structure throughout the cross section. The grain sizes of the samples shown in
Billets with diameter 95 mm were cast in a lab casting facility using the Hycast hot-top gas-slip technology without using a grain refiner. A picture of a macro etched billet slice is shown in
The cast billets were homogenized at 575° C. for 2 hours and 15 minutes followed by cooling at a rate of approximately 400° C. per hour. Micrographs of the particle structure in the billets from the two different alloys in examples 1 and 2 are shown in
The billets were extruded at an 8 MN laboratory extrusion press with a 100 mm diameter container to a profile with a cross section of 5×40 mm2. The billet preheating temperature was approximately 500° C. and the extrusion speed 20 m/min. After extrusion the profile was quenched in water.
A 100 cm long piece from the back part of the extruded profile was cold rolled to give 20, 30, 40 and 50% reduction in the thickness. The samples that were cold rolled to different thicknesses were then annealed in a salt bath which had been preheated to 500° C. A hole was drilled into each of the samples to fit a thermocouple. When a sample was put into the salt bath the holding time of 10 seconds started when the temperature reached 490° C. After annealing the samples were quenched in water. In addition one sample of the material cold rolled to 40% reduction in thickness was held 5 minutes at 500° C. Yet another sample of the material cold rolled to 40% reduction in thickness was heated in an air circulating oven at a considerably lower heating rate to the annealing temperature than that obtained in a salt bath.
A micrograph of the as extruded sample is shown in
The sample that was cold rolled to 40% reduction in thickness and held at 500° C. for 5 minutes did not show any grain growth. The reason for this is probably that the number of AlFeSi-particles is high enough to prevent grain growth. With even lower Fe contents than 0.09 wt % a holding time of 5 minutes at this temperature could cause grain growth in the sample.
Example 2 shows that it is beneficial to heat the cold rolled sample fast to the solutionizing temperature to obtain an even grain size and that a holding time of only 10 seconds is sufficient to obtain a fully recrystallized grain structure.
Example 2 also shows that the final grain structure in the blanks could be perfect for providing attractive high gloss anodized surfaces even though the billet grain structure is regarded as being far from optimum when it is cast without grain refiner.
Billets with diameter 95 mm of a 7030 alloy were cast in a lab casting facility using the Hycast hot-top gas-slip technology and a TiB2 based grain refiner. The chemical composition of the alloy is shown in Table 5.
The billets of the 7030 alloy were homogenised for 4 hours at 500° C. The billets were extruded at an 8 MN laboratory extrusion press with a 100 mm diameter container to a profile with a cross section of 5×40 mm2. The billet preheating temperature was approximately 500° C. and the extrusion speed 12.5 m/min. After extrusion the profile was quenched in water.
A 100 cm long piece from the extruded profile was cold rolled to give 20, 30, 40 and 50% reduction in the thickness.
The cold rolled samples at different thicknesses were then, one by one, put into a salt bath that had been preheated to 500° C. With a thermocouple drilled into each sample it was possible to monitor the temperature of the sample. All samples were held approximately 10 seconds at a temperature above 495° C. before quenching in water. The heating rates of the samples depended on the thickness, but in all cases the heating time was less than 10 seconds.
As shown in
The grain sizes of some of the samples depend on the depth below the surface of the cross section. Some samples have very coarse grains in the middle of the cross section and finer grains towards both surfaces. A typical machining depth to remove the coarse surface grain layer in a small profile like this would be around 250 μm, and this depth was chosen for the grain size measurements. In Table 6 (below) the grain sizes for the alloy of Example 3 are listed. By looking at these grain size measurements alone, all the samples seem to fulfil the requirement of a grain size below approximately 70 μm.
However, the pictures in
The main benefit of the present invention is that it is possible to obtain a grain structure with an even grain size and a close to random texture throughout the cross section of a profile irrespective of the grain size in the profile after it has been extruded and thus also irrespective of the grain structure of the billet before extrusion. This improvement in grain structure is obtained by cold rolling deformation of the extruded profile followed by flash annealing.
An extruded thick walled flat profile will in most cases have a coarse grain layer that according to the state of the art has to be removed in order to obtain a smooth anodized surface with a minimum of defects in the final product. The amount of material that would have to be removed in the as extruded cross section is typically in the range 7-15%.
Moreover, the cold rolling will ensure a very accurate thickness and flatness of the profile, and for that reason considerably reduce the need for machining. An extruded profile will have much more variation in the thickness, typically ±0.15 mm (the variation could also be higher for very wide profiles, especially for 7xxx alloys).
Since the grain size in the billet and the extruded profile is of little importance for the resulting grain size in the cold rolled and annealed blanks there is a possibility of casting the billets with a minimum or even completely without the use of a grain refiner. In order to avoid center cracks in the billets in the startup of the cast it could be beneficial to add some grain refiner in the first metal to cast. The grain refiner itself could be a source for inclusions that can cause failures in the anodized surface. Another benefit of not using a grain refiner is that the melt cleaning with the use of ceramic foam filters will be more effective on other type of inclusions (Nicholas Towsey, Wolfgang Schneider and Hans-Peter Krug, A comprehensive study of ceramic foam filtration, 7th Australasian Asian Pacific Course & Conference, Aluminium Cast House Technology: Theory & Practice, P. Whiteley and J. Grandfield (TMS: 2001)
The possibility of reducing the Fe content and still obtain an adequate grain structure will significantly improve with the use of the present invention. The lower Fe content can either be used to improve the gloss, or to keep the current gloss but add a thicker and more wear resistant oxide layer to the anodized product. The latter will make the product more durable.
Even though there is extra cost associated with the cold rolling and annealing process to obtain the uniform and random grain structure, this will probably be more than compensated for by the savings due to reduced machining and reduced material consumption.
Number | Date | Country | Kind |
---|---|---|---|
20151232 | Sep 2015 | NO | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/NO2016/000014 | 4/20/2016 | WO | 00 |