Background Art At present, it is difficult to use secondary aluminum-silicon alloys in industry, because of the unsatisfactory properties of these alloys. It would be desirable to use recycled alloys in various fields, for example the automotive industry, agriculture (i.e. tractors) and aviation. Recycling is a preferred alternative in industry, offering economic advantage as well as better use of natural resources.

Another important benefit of recycling is the preservation of the environment, since it uses waste metals which would otherwise contaminate the environment, and present a problem to dispose of.

Throughout the present disclosure, the term "silumin" is used to indicate an Al-Si alloy.

The problem, however, is that recycling usually results in secondary alloys, that is alloys which include more than about 0.5 % impurities, like Fe, Mg, Cu, Cr, Ni, Zn, Mn and/or others. Likewise, there are lower grade ores which result in secondary alloys. In many cases, the impurities include an enhanced percentage of iron Fe (for example, up to 0.7 %).

The problem exists not only with secondary alloys. Even a smaller percentage of the above impurities in Al-Si alloys may have a detrimental influence on the properties of the alloy.

The low performance relates to the casting and mechanical properties of these alloys, for example their castability, porosity, machinability, ductility and fatigue strength.

The undesirable properties of these secondary alloys mainly result from the structure into which these alloys solidify, with the iron contents crystallizing into specific structures which include for example a long needle morphology, or Chinese script, or needles. Structures including these morphologies with their undesirable properties appear while cooling the alloy, for example in sand cast, at cooling rates between 0.1 and 1.0 degrees K/ second. Similar structures with these morphologies also appear in other casting methods.

There are methods known in the art to partly remedy these problems.

The problem of large iron-bearing constituents can be remedied with: 1. Increasing the cooling rate 2. Reducing the iron contents 3. Adding elements which transform the iron constituents into a harmless shape 4. Adding elements to the liquid, which dissolve the undesired phases into smaller parts.

The undesired influence of Silicon impurities can be reduced with: 1. Increasing the cooling rate 2. Heat treating to dissolve or spherodize the compounds Increasing the cooling rate will work for both problems, but the required substantial increase in the cooling rate which can only be achieved by changing to other casting methods, that is from sand casting to metallic mold casting. These methods, however, tend to increase the internal tension in the casting, which may produce warping or cracking of the casting.

Metallic mold are very expensive, therefore are not preferred over sand molds.

Moreover, changing to metallic molds requires a drastic change in the method of production, a costly alternative.

A reasonably efficient and effective method is the addition of elements which transform the iron compound from relatively large plates or needles to smaller and less embrittling forms. The elements used, known as iron correctors, include manganese, chromium, nickel, cobalt, molybdenum and other elements.

These elements form compounds with the iron in the alloy, which crystallize into a phase with various forms like globular or dendritic forms, which do not have the undesired properties of the abovementioned plates and needles.

Despite their deficiencies, in many cases the addition of iron correctors is chosen as a reasonable compromise.

These additional elements, however, add a significant amount to the cost of the alloy thus formed. Moreover, these additions influence the properties of the alloy.

Thus, the percentage of iron-bearing compounds is increased to the point that they influence the solidification mode and reduce the fluidity of the alloy.

There is a complication in the manufacturing process, since the preparation of a master alloy is required. This makes the process more expensive. Master alloys are required since aluminum will not accept certain materials, molybdenum for instance.

Machinabiiity, too, is reduced, especially if primary crystals of the compounds are formed.

As known in the art, the silicon phase can be present in several structures. The eutectic can be random, not-modified, undermodified, modified and overmodified.

The primary silicon crystals can appear as globular or plate-like shapes, as well as feathery, starshaped or spherodized.

If the alloy contains more than 0.8% Fe, then primary Fe Si Al5 crystals appear.

If Mn is also present in the alloy, then the compound (Fe Mn)3 Si2 A115 is formed.

This compound has the shape of Chinese script, thus the embrittling effect of Fe Si Al5 is eliminated.

If the total manganese plus iron contents in the alloy exceeds 0.8%, then the (Fe Mn)3 Si2 Al15 crystals are primary and they appear as hexagonal globules.

These globules do not embrittle the alloy, but they reduce its machinability.

If the Cr or Ni are present in the alloy, the compounds (CrFeMn)x Siy Alz or (NiFeMn)x SiyAlz are formed, respectively.

A preferred method for decreasing the mean size of the iron-bearing precipitates of the eutectic origin includes the modification of the alloy structure by the introduction of small amounts of specific elements (modifiers). The idea is to achieve a more disperse structure of the eutectics and, as a result, to decrease the size of the eutectic iron-bearing inclusions which are comparable in size with eutectic phases.

All the alkaline and most of the alkaline earth metals achieve the modification effect.

The most used metal is sodium which is also the cheapest. Strontium is also widely used. The other alkaline earth metals are less effective.

To achieve the desired modification effect, the percentage of the sodium addition should be about 0.01 - 0.02 % Because of the limited miscibility and the strong tendency of sodium to oxidize, however, larger amount of sodium are added, especially if the melt is not poured immediately. Sodium is a difficult metal to handle. It tends to float on the melt and has to be kept immersed until melted; it oxidizes rapidly and its effect disappears in a short time.

It is known that sodium decreases the grain size of the alloy.

This is a desirable effect. Sodium, however has an adverse effect on the castability of the melt. Sodium has no influence on the iron phase, therefore is more useful for relatively clean alloys.

Modification can also be produced with alkaline metal salts if they decompose in contact with the melt, but the salts which are effective are very expensive and are not too efficient.

The elements which nucleate the silicon and distribute the primary silicon crystals include: arsenic, sulfur, selenium, tellurium and gallium plus tellurium. Boron together with titanium refines the grain size of the aluminum hut does not appreciably affect the silicon appearance.

Small amounts of alkaline or alkaline earth metals or alkaline metal salts change drastically the appearance of the silicon crystals, which become smaller, more rounded and for a coupled eutectic. Basically, there are two processes which occur: 1. The eutectic changes from separated to coupled 2. There is a decrease in the surface tension of aluminum, that leads to silicon particles which are more rounded and smaller.

Unfortunately, there is no transformation of the iron-bearing constituents, which retain their undesired shape.

One of the most efficient methods to decrease the volume fraction and the main size of the primary iron-bearing precipitates is sufficient overheating of the silumin melt, above liquidus. It was found that liquid Al-Si alloys conserve microheterogeneous state for a long time duration after the ingot melting or components mixing at a temperature above liquidus. The microheterogeneity is inherited from the initial heterogeneous property of the material.

Just after their melting, fragments of various solid phases begin to solve. However, the dissolving process is not completed immediately, and no true solution is formed in the initial stage.

During a first stage, the melt has a colloidal structure, comprising an aluminum-bearing solvent and disperse (of the order of 10 nm size) colloidal particles including silicon, iron and other elements.

These particles either dissolve very slowly or remain in a metastable state of equilibrium with the surrounding melt.

In any case, at a temperature slightly above liquid us, the system conserves its microheterogeneity for a time period of the order of 10 hours, that is during the whole melting process. When the melt is crystallized, the above-mentioned colloidal particles become the nuclei of solid silicon-, iron- and other elements- bearing phases. This may result in an alloy with inferior performance, as detailed above.

The above processes are detailed in literature as follows: 1. L.F. Mondolfo, "Aluminum Alloys: Structure and Properties". 1979 p. 971.

2. 1. Minkoff, "Solidification and Cast Structure" . 1986. John Wiley and Sons.

3. Naeker, "Proceedings of the Conference on Thermal Analysis of Molten Aluminium" .1985. P.155.

4. Zhao et al., "Effect of Zn on microstructure and properties of Al-Si alloy" Taiyuan Univ. of Technology/.Journal of Special Casting & Nonferrous Alloys. 2 1994. pp. 5-7 . Language: Chinese.

The paper deals with the effect of Zn on the structure and properties of Al-Si alloy.

Eutectic cell structure is formed in the modified alloy by the addition of certain Zn.

5. Chichko et al., "On the parameters of nucleation of modified and nonmodified silumin metals" . Belorusskaya Gosudarstvennaya Politekhnicheskaya Akademiya, Minsk, Beloruus. Rasplavy n 5 Sep-Oct 1993, pp 83-86. Language: Russian.

Cooling curves are built for non-inoculated and Na-inoculated aluminum-11.5 - 12.5% silicon alloys. The nuclei number and growth rate are estimated as a function of temperature (35-125 degree C) for each curve by the program developed. The Kolmogorov model corrected for the heat balance equation.

Patents in prior art which may have a relationship to the present invention include the following: LANGENBECK et al., US Patent 4799978, Patent Assignee: LOCKHEED AIRCRAFT CORP., details an aluminium powder alloy having good high temperature performance - contains iron, nickel and chromium.

Hot worked Al alloy powder article consists of (in%): al 81-91.9 esp. 86, Ni 4-8 esp. 6, Fe 4-8 esp. tand Cr 0.1-3 esp. 2.

USE/ADVANTAGE - Esp. in mfr. of aircraft parts exposed to elevated temps.

Alloy retains it mechanical properties even after prolonged exposure to temps. up to 800 deg.F.

MAHAJAN et al., US Patent 4787943, Patent Assignee: (USAF ) US SEC OF AIR FORCE, details a dispersion strengthened aluminium alloy - contains titanium and rare earth(s). Al alloy dispersion-strengthened with rare earth metal(s) comprises (in wt.%): Ti 2-6, rare earth(s) 3-11, (VIII) element(s) 3 max and Al the balance.

Pref. rare earth is Gd. Pref. max. at ratio of Ti to rare earth is about 2:1. The amt. of (VIII) element is 0.1-3.0 wt.% and the pref. element is Fe. A specific alloy has the compsn. Al-4Ti-4Gd. In a typical process, 75 micron thick ribbon is formed by casting onto a chill wheel and annealing at 100-600 deg.C for about 1 hr.

SCHUSTER et al., WO 8706624, US 4786467, Patent Assignee: (ALCN ) ALCAN INT LTD, details a cast metallic matrix with refractory reinforcement - with good stiffness obtd. by mixing melt without gas entrainment using shearing action.

In the prodn. of a composite material comprising an alloy reinforced with particulate non-metallic refractory the latter is mixed with the molten matrix under such conditions that introduction and retention of gas are minimised, and that the particles do not degrade in the mixing time. In mixing the particles and melt are sheared past each other to promote wetting.

The mixture is then cast at such a temp. that no solid metal is present. Pref. the matrix phase is an aluminium alloy and the reinforcing phase is silicon carbide, alumina, boron carbide, boron nitride or silicon nitride.

KUBO metal., EP 241198, JP 62240727, US 4789605, Patent Assignee: (TOYS) TOYOTA JIDOSHA KK, details a light metal matrix composite material with good high temp. properties - having reinforcing phase including potassium titanate whiskers.

A composite material comprises a matrix of light metal reinforced with a mixture of potassium titan ate whiskers and short fibre material selected from silicon carbide or nitride whiskers, alumina short fibres, crystalline or amorphcus alumina-silica short fibres. The overall proportion of reinforcing phase is 5 to 50 vol.%, and the propn. of titanate whiskers in that phase is 10 to 80 vol.%. Pref. the propn. of reinforcing phase is 10-40% and the propn. of titanate in it 20-60%.

Thus, the above patents and literature do not disclose an approach comprising both modification and overheating. There are patents using sodium, but no patents were found which disclose the use of strontium.

Recycling of metals is highly desirable in the automotive industry, as well as in agriculture (i.e. tractors) and aviation. A problem with recycling is that it usually results in secondary aluminum-silicon alloys, that is alloys which include impurities such as Fe, Mg, Cu, Zn, Cr, Mn and/or Ni. These secondary alloys have unsatisfactory properties, for example mechanical properties and poor castability, as well as unsatisfactory heat resistance and other properties.

These unsatisfactory properties mainly stem from the structure into which these alloys solidify, for example Fe precipitating into long needles, plates or skeleton morphology shapes (about 100 micrometers long). Prior art solutions to the iron-rich precipitates include the addition of elements known as "iron correctors". These additives include manganese, chromium, nickel, cobalt, molybdenum and others. They form compounds with the iron in the alloy, which crystallize into various forms that do not have the abovementioned undesired properties of the prior plate or needle structures.

For example, if Mn is introduced into the alloy, then the compound (FeMn)x Siy Alz is formed, which compound has the shape of Chinese script (skeletons), thus eliminating the embrittling effect of Fe Si Al5.

The above additional elements, however, significantly increase the cost of the alloy.

Moreover, other properties are detrimentally affected, including a reduction in the fluidity of the ailoy, and a reduction in machinability. It is known that sodium is a good modifier, however it has a detrimental effect on castability.

It is an objective of the present invention to provide for aluminum alloys and methods for their production with means for overcoming the abovedetailed deficiencies.

Disclosure of Invention It is an object of the present invention to provide Al-Si casting alloys.

A modifier improves the properties of Al-Si alloys, and is especially useful for alloys with impurities, alloys which at present have inferior mechanical properties, for example secondary Al-Si alloys. Secondary alloys usually refers to alloys which include more than about 0.5 % impurities like Fe, Mg, Cu, Cr, Ni and others. The resulting secondary Al-Si alloy (or silumin) has improved properties, as detailed below.

This object is achieved by Al-Si casting alloys as disclosed in claim 1.

In accordance with the invention, the object is basically accomplished with the addition of modifiers in the form of ultrafine powders, which remain in solid state in the melt and form the nuclei around which iron solidifies into forms having the desired properties. Throughout the present disclosure, "powders" refer to ultrafine powders.

Moreover, the modifiers are ultradisperse powders made of TiN or AIN which materials it was found to result in the alloys with the desired properties. Other powders which contain carbides and/or nitrides and/or carbonitrides may be used, provided that these powders can act as nucleant for the iron phases.

It is another object of the present invention to provide a method for the production of the alloy which includes the process of overheating the melt above the homogenization temperature, prior to casting.

According to the present invention, the method of production of the alloy includes a combination of overheating the liquid melt and the addition of the modifier powder prior to casting.

According to another aspect of the present invention, the powder may be added to the melt either directly or within a master alloy. The preparation of a master alloy is preferred where the powder has a specific density which is different than that of the melt.

A master alloy preferably also contains Al, Mg and/or Cu, in addition to the ultrafine powder.

According to a still another aspect of the present invention, the preferred concentration of modifier (by weight) should represent about 0.01% to 0.019% of the mass of the melt.

The method may include overheating, modifier addition and casting. Ultrasonic treatment of the melt may be performed either before or after modification.

Further objects, advantages and other features of the present invention will become obvious to those skilled in the art upon reading the disclosure set forth hereinafter.

Brief Description of Drawings The invention will now be described by way of example and with reference to the accompanying drawings in which: Fig. 1 illustrates the temperature dependence for the viscosity of the liquid silumin, during the heating stage and the subsequent cooling stage.

Fig. 2 illustrates the temperature dependence for the density of the liquid silumin, during the heating stage and the subsequent cooling stage.

Fig. 3 - A photograph illustrating the microstructure in prior art secondary alloy.

Fig. 4 - A photograph illustrating the microstructure in alloy made using the new method.

Modes for Carrying out the Invention A preferred embodiment of the present invention will now be described by way of example and with reference to the accompanying drawings.

The invention discloses Al-Si casting alloys and methods for their production using modification with an ultrafine powder. The methods improve the properties of silumin alloys with impurities, and are especially suitable for secondary alloys.

Method for modification using ultrafine powders A method for producing an Al-Si casting alloy comprises the following steps: 1. Heat the alloy to a temperature which is about Delta~T degrees C above the liquidus temperature; 2. add a modifier comprising an ultradisperse powder capable of remaining in solid state during all the stages of the alloy preparation process; 3. degassing (or fluxing) of the alloy; and 4. pour the melt.

To achieve the desired properties, the modification has to conform to the following conditions: 1. The lattice conformity between alpha-Al and the powder nuclei; 2. Synthetic crystals must be provided, which can be successfully added to the liquid silumin, either directly or contained in a master alloy; 3. The pulverized powder must be a stable phase within liquid silumin, including at the elevated overheating temperature; 4. A sufficient dispersiveness of the powders must be achieved, so that nuclei to iron-bearing constituents be dispersed throughout the melt.

According to the present invention, it was found that the addition of synthetic titanium nitride (TiN) powder to the liquid silumin having an enhanced iron contents results in an alloy with improved properties.

TiN is a stable phase in the liquid silumin, prior to casting.

Thus, TiN powder particles each becomes a center of crystallization for phases which contain iron Fe, and it reduces the iron phases to fragments.

For example, in an alloy containing 6.98% Si, 1.2% Fe, 0.35% Mg, 0.31% Mn and other impurities, it was found that the needles in the initial aluminum alloy were about 50 to 150 microns long.

After the addition of the modifier powder (with the nucleation process), it was found that the length of needles in the alloy is about 20-70 microns, the smaller length corresponding to improved mechanical properties.

Moreover, limiting the silicon plates growth by TiN leads to change of eutectics kind based on silicon.

This is a novel aspect of the present invention - the use of synthetic powders as nuclei for secondary aluminum alloys.

TiN is just one of many types of ultradisperse powders which may be used as nucieation centers in silumin alloys. Other powders may be made of AIN, carbides and/or nitrides and/or carbonitrides. It is also possible to use a mixture of the above powders, to achieve still better performance.

One requirement of these powders is their capability to act as nucleant for the iron phases. It was found that materials with a cubic lattice form are suitable as nucleants. Some materials with hexagonal lattice may be suitable as well.

Another requirement of these powders is that they remain in solid state in the melt at the liquid temperature. If an overheating process is used and the modification is made prior to overheating, then the powders have to remain solid even at the overheating temperature. Overheating temperatures may be about 1100 degrees Celsius, or 1200 degrees, or up to 1400 degrees, according to the specific composition of the alloy.

Whereas heterogeneous nucleation can be regarded as a geometrically modified case of homogenous nucleation by which the activation barrier is decreased by the presence of a foreign substrate, a calcuiation from the geometrical point of view shows that, when the liquid/soiid interface of the substance is partly replaced by an area of low energy solid/solid interface between the crystal and a foreign solid, then nucleation can be greatly facilitated.

Normally, the potentia! of the foreign substrate in facilitating the nucleation process is estimated from the following equation: f(Q) = (2+cos(Q))(1-cos(Q))/4 where Q is the wetting angle between the growing crystal and the foreign substrate within the melt. Under conditions of good solid/solid wetting, that is small Q, the foreign substrate can have a dramatic effect on the nucleation process.

Thus, the silumin alloys include modifiers in the form of ultrafine powders, which remain in solid state in the melt and form the nuclei around which iron solidifies into forms having the desired properties.

Accordingly, pulverized powder of TiN may be used as a modifier.

An ultradisperse powder TiN was used, of dimensions 20 to 80 nm (nanometers). It was found that, as the size of the powder particles is smaller, the properties of the resulting alloy are better.

Thus, it is preferable to use smaller size particles. The powder size detailed above corresponds to presently available powders. As smaller powders become available (for example in the 10 nm range) it is recommended to use these finer powders where improved performance is required.

This is a synthetic powder of cubic crystalline shape.

The TiN used has a lattice parameter of a=4.24173 (space group Fm3m) while the lattice parameter of aluminum decreases linearly, to reach a value of a=4.478 E-10 meter at 1% Si and a= 4.0365E-10 meter at 11%.

The TiN powder modifier, in the form of a master alloy, was introduced into the silumin melt at a temperature of 720 to 750 degrees C.

The temperature is determined from the phases diagram (Al-Si), and varies according to the relative percentage of Si. In the following description, where a specific temperature is mentioned for clarity, it is to be understood that it is the temperature derived as above detailed.

The composition of the master alloy included 7% TiN, 7% Mg and 86% Al.

It was found that good performance is achieved with master alloys including about x% of Mg or Cu or a combination thereof, and about x% of TiN powder, with the rest Al, and wherein x is in the range about 3 to 9 percent.

In other embodiments, still better results were achieved using a master alloy with x about 7 percent.

In other experiments, it was found that the desirable alloy properties were achieved with the addition of the modifier into the alloy at temperature of about 720 and up to 1000 degrees C.

According to the present invention, it was found that overheating the melt prior to casting may be used to further improve the properties of the alloy.

Possible causes for the inferior performance in prior art methods were investigated at AMT Ltd. The cause may be the colloidal structure of the melt, comprising an aluminum-bearing solvent and disperse (of the order of 10 nm size) colloidal particles including silicon, iron and other elements.

At a temperature slightly above liquidus, the system conserves its microheterogeneity for a time period of the order of 10 hours, that is during the whole melting process. When the melt is crystallized, the above-mentioned colloidal particles become the nuclei of solid silicon-, iron- and other elements-bearing phases.

Scientists at AMT Ltd. have found that the microheterogeneous state of liquid silumin can be irreversibly destroyed by overheating the melt to a higher temperature above liquidus.

To achieve this effect, it is required to heat the silumin alloy to a temperature of about 1000 degrees C and to keep it at that temperature for a predefined time period, for example more than 10 minutes.

The specific temperature of overheating depends on the composition of the melt. The overheating process transforms the melt to a true solution.

A method of silumin production thus includes the following steps: 1. Heat the silumin alloy to a temperature Tp which is Delta~T degrees C above the liquidus temperature. If was found that Delta~T in the range of 20 to 40 degrees C achieves the desired effect. The required Delta~T depends on the specific composition of each alloy being considered.

Usually, the temperature is in the range 720 to 750 degrees C.

Tp refers to the pouring temperature of the melt; 2. add the modifier, the ultradisperse powder TiN, of dimensions of 10 to 80 nm (nanometer). In a preferred embodiment, the quantity of powder is within the range about 0.01% to 0.019% of the melt, by weight; 3. overheat the melt to about 1000 degrees, and keep the alloy at that temperature for about 20 minutes. The required overheating temperature depends on the structure of the alloy, and may range between 1000 and 1400 degrees C.

This is a temperature where the melt becomes homogeneous; 4. cool the melt back to the pouring temperature Tp, for example 750 degrees C, or a value suitable for that specific alloy; 5. degassing (or fluxing) using a procedure as known in the art, for example using Ar or salts; 6. pour the melt. A sand cast is preferably used.

The melt should be preferably overheated above its homogenization temperature (or above the temperature of its property-temperature dependence branching), and the alloy kept at that temperature for about 20 - 30 minutes. This achieves a homogeneous alloy.

For best results, while using sand casting the following points should be watched: gas contents, grain refinement and modification of the aluminum silicon alloys.

In another embodiment of the present invention, the method of production includes the steps 1 - 5 as detailed above, but in a modified order: 1. Overheat the melt to 1000 degrees, and keep the alloy at that temperature for about 20 minutes. Again, the exact temperature depends on the specific alloy, and may be between about 1000 to 1400 degrees C; 2. cool the melt back to 720 -750 degrees (actually to a temperature suitable to the specific alloy being processed, here referred to as the pouring temperature Tp); 3. degassing (or fluxing) using a procedure as known in the art, for example using salts or Ar; 4. add the modifier, the ultradisperse powder TiN, of dimensions of 10 to 80 nm (nanometer), in a quantity within the range about 0.01% to 0.019% of the melt, by weight; 5. pour the melt into sand cast.

This method was also found to achieve good results.

The above methods were found to give satisfactory results with sand casting, which is the usual, low cost method in use. Thus, more complex methods, which also create new problems as detailed above, are avoided.

It was found that, since the TiN powder has a specific density which is different than that of the silumin, that a master alloy is required.

Thus, it is recommended that TiN be added as a master alloy, rather than directly as a powder. This ensures that the powder will be uniformly dispersed in the liquid alloy, to achieve a generally uniform casting with the desired properties.

In other powders, like AIN, which have a specific density similar to that of the silumin, no master alloy is required, and the powder may be added directly to the melt. According to the invention, the powder was not added directly, but was encapsulated in a piece of aluminum foil.

Thus an ampoule was obtained.

This structure ensures the powder will penetrate into the melt, and the foil will dissolve in the aluminum alloy with no impurities added.

Other ultradisperse powders may be used as modifiers, for example carbides or nitrides or carbonitrides, having dimensions in the range of about 0.01 to 0.08 microns.

In another embodiment of the invention, a mixture of powders is used for still higher performance.

Thus, the modifier powder comprises a combination of ultradisperse powder of AIN and/or TiN and/or carbides and/or nitrides and/or carbonitrides, each powder having dimensions in the range of about 0.01 to 0.08 microns.

A preferred embodiment of the invention The methods II and Ill detailed below were found to achieve alloys with superior performance. Method IV also achieves good performance. Comparative tests were performed on alloys prepared using the novel casting method, whose performance was evaluated against alloys prepared according to the prior art casting method, as detailed below.

For the experiments performed to test the present invention, an aluminum casting alloy was prepared including: 6.98% Si, 1.12% Fe, 0.35% Mg, 0.31% Mn, 0.18% Cu, 0.03% Cr, 0.17% Ti, 0.36% Zn, 0.01% Ni. This is but one example of secondary aluminum alloy compositions which can be improved according to the present invention.

Casting Method I (Prior art) The method includes the following steps: A. Heating the alloy to about 720 degrees C B. degassing for 10 minutes C. casting in a dry-sand mold Casting Method II (New method) The method includes the following steps: A. Heating the alloy to the overheating temperature, Toh. In the tests performed at AMT Ltd., Toh was about 1100 degrees Celsius; B. holding for about 30 min. (minutes) C. cooling the melt together with the furnace, to the pouring temperature Tp. In tests performed at AMT Ltd., Tp was about 720 degrees Celsius; D. degassing for about 10 min.; E. addition of TiN powder within a master alloy to the melt, about 0.015 % powder by weight; F. holding for about 3 min., while mixing the melt; G. casting into a dry-sand mold.

Degassing was performed using hexachlorethan as known in the art.

Other degassing materials may be used, for example argon.

A synthetic titanium nitride TiN ultradisperse powder was used, with dimensions of about 20 to 70 nanometer, to prepare the master alloy.

The powder within the master alloy was added directly into the liquid aluminum alloy at the pouring temperature Tp of about 720 - 730 deg.C.

Casting Method III (New method) The method includes the following steps: A. Heating the alloy to the overheating temperature, Toh. In the tests performed at AMT, Toh was about 1100 degrees Celsius; B. holding for about 30 min.; C. cooling the melt together with the furnace, to the pouring temperature Tp. In tests performed at AMT Ltd., Tp was about 720 degrees Celsius; D. degassing for about 10 min.; E. addition of AIN powder to the melt, about 0.015 % powder by weight; F. holding for about 3 min., with mixing of the melt; G. casting into a dry-sand mold.

A synthetic aluminum nitride AIN ultradisperse powder was used, with dimensions of about 20 to 70 nanometer.

The powder was added directly into the liquid aluminum alloy at the pouring temperature Tp of about 720 deg.C.

Casting method IV (New method) The method includes the following steps: A. Heating the alloy to the overheating temperature, Toh. In the tests performed at AMT, Toh was about 1100 degrees Celsius.

B. holding for about 30 min.

C. cooling the melt together with the furnace, to the pouring temperature Tp. In tests performed at AMT Ltd., Tp was about 720 degrees Celsius.

D. degassing for about 10 min.

E. addition of a mixture of TiN powder and AIN powder to the melt, about 0.015 % total powder by weight F. holding for about 3 min., with continuous mixing of the melt G. casting into a dry-sand mold Test methodology and results The microstructure and composition of the specimens were evaluated by optical microscopy and Scanning Electron Microscopy (SEM) equipped with the Energy Dispersive Spectroscopy (EDS). (Link Oxford) was mounted on the SEM. SEM was conducted on a JEOL 840SEM. The same specimen was used for microhardness testing routinely operated at 10 kV and 20 kV.

TEM was conducted on a JEOL 2000FX, routinely operated at 200 kV. The 2000FX also includes a link EDS system which was used for micro-elemental analysis.

The initial aluminum-silicon alloy was prepared from ingots by prior art Casting method I. The alloy used included Al, about 13% Si, 1.09% Fe and other impurities.

The random eutectic globular primary silicon, elongated phases of (CrFe)4 Si4 Al13 and FeNiAls, primary crystals of (FeMn)3 Si2 Alias, the (Cu, Ni)2 Al3 phase in the form of Chinese script and small needles, small quantities of FeSiAls and (CuNi)2 Als (light small scripts) were clearly visible on the surface of the solid. The size of these structural compounds are presented in Table 1. The E.G. initial Vickers hardness of the matrix was found to be about 86.8.

The modified aluminum-silicon alloys according to the present invention were prepared from ingots by Casting methods II and III.

The finely dispersed eutectic and dispersed silicon were clearly visible in both cases of modification.

For ingots prepared according to new Method II, the TiN powder particles acted as nucleation centers for the iron containing phases such as (CrFe)4 Si4 AIls Fe Ni Alps, (FeMn)3 Si2 AIls and Fe Si Alps. As a result, those phases were fragmented. In addition, after overheating, a redistribution of Fe, Cr, Ni and Mn in these phases was observed. Thus, TiN powder is believed to act primarily as a nucleant.

Whereas the initial Vickers hardness of the matrix (Method I) was about 70, the modified alloy with titanium nitride (Method II) had an improved hardness of about 105.5.

For ingots prepared according to new Method ill, the AIN powder particles acted as modificant and nucleation centers mainly for the aluminum matrix.

The iron - enriched phases were found to be located on the grain boundaries of the typically modified alpha-Al phase.

A redistribution of Cu, Mg, Fe, Si, Zn, Mn, Cr and Ni as a result of overheating and modification with AIN powder was observed.

Whereas the initial Vickers hardness of the matrix (Method I) was about 70, the modified alloy with aluminum nitride (Method III) had an improved hardness of about 103.9.

Table 1 details the size of the various phases after casting using the methods l, ll and III. The improvement (the decrease in size of these phases) is evident.

Table 1. The size of the various phases in the cast Phase Size, micrometer Method I Method II Method III Prior art Novel, TiN Novel, AIN Si 10-30 5-10 5-10 (globular) (eutectic) (eutectic) 40-60 (eutectic) (CrFe)4 Si4 Al13 40-120 10-40 30-80 FeNiAls 40-60 5-25 15-25 (FeMn)3 Si2 AIls 60-80 25-35 40-45 FeSiAls 25-30 10-15 20-25 CuSi4 Mg8 Al4 20-60 25-60 25-60 (CuNi)2Als 25 25 25 * For Al alloy with 13.1% Si and other impurities Thus, it was found that both novel Methods II and Ill improved the castability of the melt by about 17%. The improved castability was measured using a castability test with castability test molds as known in the art.

Therefore, the abovedetailed methods can be applied for recycling of silumines with impurities. The methods may be also applied to eutectic, hypoeutectic and/or hypereutectic secondary silumines.

According to the present invention, synthetic AIN or other powders with a specific density similar to that of silumin may be added directly to the liquid silumin, to achieve an alloy with a fine microstructure.

If a synthetic powder having a specific density which is different from that of silumin is used, for example TiN, then it is recommended that first a master alloy be prepared. The master alloy is then added to the liquid silumin, to achieve an alloy having a fine microstructure.

In either case, this results in fine iron-enriched phases, finely dispersed eutectic and dispersed silicon.

Various embodiments of the present invention are possible.

For example, although the preferred quantity of the TiN and/or AIN powder to be added to the melt is about 0.015 % powder by weight, it is possible to add powder in the range 0.012 % to 0.018 % by weight to achieve the desired abovedetailed effects. In another embodiment, the powder concentration was up to 1% by weight, and the desired improvements were still achieved.

A mixture of AIN powder and TiN powder according to the abovedetailed Method IV may be advantageous for achieving still smaller size phases.

It appears that the two powders may have different effects, the TiN acting more as nucleant for the Fe phases, whereas the AIN acts more as a modifier and nucleant for the aluminum phases. Thus, a synergetic effect may be achieved by using a mixture of the two powders, with respect to the size of the phases.

Where the castability properties are more important, then the TiN powder alone may achieve better results.

The preferred size of the TiN and/or AIN powder addition is between 10 and 80 nanometer, that is 0.01 to 0.08 microns. The smaller the size of the powder, the better the properties of the alloy. Smaller size powders, however, demand a higher cost. Therefore, there is a tradeoff between cost of additive and the resulting properties of the silumin alloy.

The minimum size, 20 nanometer, is the finest powder which is commercially available. When finer powders become available, it is recommended to use them for improved performance. Powder larger than 80 nanometer are not as effective as nucleant/modifier.

The pouring temperature Tp was about 720 degrees Celsius during tests performed at AMT Ltd. It is possible to hold at higher or lower temperatures. A pouring temperature Tp in the range 700 to 740 degrees Celsius is preferred, and according to the composition of the alloy.

The AIN and/or TiN powder are preferably added to the melt after overheating and reducing the temperature to about 720 to 740 degrees C.

Whereas the melt is heterogeneous prior to overheating, it becomes homogeneous as a result of overheating. It is better to add the powder to the homogeneous melt. Still, it is possible to add the powder prior to- or during overheating. The powder is not affected by overheating.

The powder may also be added to the alloy in a process which does not include overheating at all. In this case, however, inferior results are to be expected. The alloy remains microheterogeneous, with a colloidal structure including aluminum-bearing solvent and disperse (size of about 10 nm) colloidal particles enriched with silicon, iron and other elements. These particles either dissolve very slowly or achieve a metastable equilibrium with the surrounding melt. In any case, at temperatures slightly above liquefaction, the melt may conserve its microheterogeneity for a time of about 10 hours, that is for all practical purposes during the whole period of the melting process.

When the abovedetailed microheterogeneous melt solidifies, the colloidal particles therein become the nuclei of solid silicon phases as well as iron and other element-bearing phases, resulting in inferior properties.

The abovementioned microheterogeneous structure of liquid silumin can be irreversibly destroyed by overheating, as detailed with reference to Methods 11,111 and IV above.

Thus overheating achieves a true solution, in which the TiN and/or AIN powders act effectively to achieve the desired properties, including machinability and castability inter alia.

It was found that, even without the addition of the powder, overheating improves the propertic s of the alloy. Thus, two sets of samples were cast at the cooling rate of 1 K/s corresponding to sand cast. The first set was heated in liquid state up to 700 degrees C and cast at the same temperature. The second set was overheated to up to 1100 degrees C (the temperature depends on the composition of the alloy) for about 20 to 30 minutes, then cooled down to 720 degrees C for an eutectic alloy, and was cast at the same temperature.

The measured hardness of the first and second sets was found to be 770 and 970 MPa respectively. Thus, overheating of the melt results in an increase of the microhardness of the alpha-dendrites.

Moreover, it was found that overheating reduces the volume percent of iron-bearing alumides by 25-30% , and the average size of the alumides decreases down to 8 - 10 microns.

The exact temperature of overheating Toh depends on the structure of the silumin alloy. For a given batch of silumin, it is possible to perform tests on samples of the alloy to determine the required overheating temperature Toh.

The overheating temperature Toh is also referred in the present application as the homogenizing temperature.

According to the present invention, the tests include continuous measurement (or taking samples at specific intervals) of the viscosity-temperature dependence and/or the density - temperature dependence of the liquid silumin, during heating and subsequent cooling.

It was found that, as the molten alloy achieves the required overheating temperature Toh and the microheterogeneous structure of liquid silumin is irreversibly destroyed, the properties of the alloy change accordingly.

Thus, the properties of the alloy prior to reaching Toh and after reaching it are different. These findings are used in a method, according to the present invention, to determine the homogenizing temperature Toh, as follows.

Measurement of homogenizing temperature Toh. Method V A. Heating the silumin alloy to well above 1000 degrees Celsius, while measuring and recording the viscosity of the melt vs. temperature.

Measurements are taken either continuously or as a set of samples.

B. cooling the silumin alloy down, while continuing the process of measuring and recording the viscosity of the melt vs. temperature.

C. finding the divergence temperature, that is the temperature where the graph for the heating stage (step A above) departs or branches off from the graph for the cooling stage (step B above). The divergence temperature is the minimum value for the homogenizing temperature, or the overheating temperature Toh to be applied during the overheating process.

Fig. 1 illustrates the temperature dependence for the viscosity of the liquid silumin, with graph 11 indicating viscosity during the heating stage and graph 12 indicating the viscosity during the subsequent cooling stage. The solid circles indicate the cooling stage.

This corresponds to an Al alloy with 6.7% Si and impurities.

One can see the splitting of the graphs at about point 13 in the graph, at about 940 deg. C. Thus, an overheating temperature slightly above this value should be used.

Measurement of homogenizing temperature Toh. Method Vl A. Heating the silumin alloy to well above 1000 degrees Celsius, while measuring and recording the density of the melt vs. temperature.

Measurements are taken either continuously or as a set of samples; B. cooling the silumin alloy down, while continuing the process of measuring and recording the density of the melt vs. temperature; C. finding the divergence temperature, that is the temperature where the graph for the heating stage (step A above) departs or branches off from the graph for the cooling stage (step B above). The divergence temperature is the minimum value for the homogenizing temperature, or the overheating temperature Toh to be applied during the overheating process.

Fig. 2 illustrates the temperature dependence for the density of the liquid silumin (Al - 13.1% Si and impurities), with graph 21 indicating density during the heating stage and graph 22 indicating the density during the subsequent cooling stage.

One can see the splitting of the graphs at about point 23 in the graph, corresponding to a temperature of about 1000 deg.C. A different temperature value may be found in other experiments, according to the composition of the alloy.

It is possible to use a combination of the above methods V and VI, for example taking the maximum value of temperature Toh, from the results found using the two methods, Method V and Method VI.

Thus, an overheating temperature Toh of slightly above 1000 deg. C is recommended for this alloy.

Therefore, for the specific alloy under test in the presented example, the complete transition from a microheterogeneous melt transition to a true solution state is achieved at a temperature slightly above 1000 deg.C.

The chemical analysis indicates that the alloy composition stays unchanged after the melt overheating in frames of the analysis precision.

Thus, the present invention teaches of using modifiers in the form of ultrafine powders. The materials may have a specific density similar to that of the silumin alloy to be improved, in which case the powder may be added directly to the melt. Otherwise, that is in case the specific densities are dissimilar, then a master alloy is preferably prepared, then is added to the melt.

The invention teaches of criteria for selection of materials for powders: 1. The material should have a cubic crystalline structure. However, some materials with hexagonal structure are suitable as well.

Suitable powders are TiN or AIN , or materials including Mg or Cu alloy with Al, or materials containing carbides and/or nitrides and/or carbonitrides.

2. The materials should have the capability to act as nucleant for the iron phases.

3. In case that overheating is used and modification takes place prior to overheating, the material used should be capable of remaining in solid state at the overheating temperature, that is in the range of 1000 to 1400 deg. C, depending on the specific silumin alloy.

To illustrate the improvement using methods disclosed in the present invention, comparative photographs are presented.

Fig. 3 is a photograph illustrating the phases in prior art secondary alloy, whereas Fig. 4 is a photograph illustrating the phases in alloy made with new method.

One can see the dramatic reduction in the size of the Si and Fe-containing phases.

There was a reduction in the size of (Fe Cr Mn)x Siy Alz phases, as well as an improvement in their shape - it become more rounded, less needle-shaped. The new form corresponds to improved alloy properties.

Various other embodiments of the present invention are possible. For example, it is possible to use aluminum nitride AIN in lieu of the TiN detailed above, with similar results. The AIN should be in ultrafine powder form, of size 0.02 to 0.08 micron, and should be added according to the abovedetailed methods.

In other experiments, a quantity of between 0.01% and 0.015% of powder by weight was added, with good results.

In other embodiments of the invention, the quantity of powder was up to 1 % , with good results. It may be possible to achieve the desired properties for even higher modifier concentration.

It is possible to add several types of powder, for example 3 or 4 types. The powder may be formed into a cube, for example with sintering or without it. The cube may include a powder like TiN, together with Cu or Mg and Al. Thus, pre-formed cubes or cylinders or pills may be used, the size according to the quantity of melt to be improved.

It was found that ultrasonic treatment of the melt, in addition to the ultrafine powder modification, further improves the properties of the alloy. The ultrasonic treatment should be preferably applied for more than 1 minute. It may be applied before and/or after the addition of the modifier.

It will be recognized that the foregoing is but one example of an alloy and method within the scope of the present invention and that various modifications will occur to those skilled in the art upon reading the disclosure set forth hereinbefore.