Kind Code:

Die casting of some magnesium alloys may yield castings that are susceptible to corrosion when exposed to salt water or other aggressive oxidizing environments. Such corrosion may result from the existence of different microstructures in a cross-section of the die casting (e.g. between the surface of the part and the center) that produce galvanic couples that are susceptible to such corrosive attack. However, a die temperature may be determined for casting of the part such that a more uniform cross-sectional microstructure is produced in which minimal or negligible galvanic potentials are produced.

Bharadwaj, Mridula D. (Bangalore, IN)
Wang, Yar-ming (Rochester Hills, MI, US)
Powell, Bob R. (Birmingham, MI, US)
Radovic, Dusanka (Sterling Heights, MI, US)
Application Number:
Publication Date:
Filing Date:
Primary Class:
International Classes:
View Patent Images:

Primary Examiner:
Attorney, Agent or Firm:
General Motors Corporation (TROY, MI, US)
1. A method of making a cast article of a magnesium alloy composition that may yield metallurgical phases of different compositions in different sections of the article when a charge of molten alloy composition is received in a set of closed metal dies and cooled in the dies to solidify the charge and shape the article, the method comprising; predetermining a temperature or temperature range at which the set of dies may be maintained to receive the charge of molten magnesium alloy and cool and solidify the alloy at a cooling rate such that the cast microstructure does not produce a galvanic coupling for a liquid environment to which the part may be exposed; and, thereafter maintaining the temperature of the dies at the predetermined temperature for casting of such parts.

2. A method as recited in claim 1 in which the temperature for the set of dies is predetermined by producing castings at different test temperatures until a useful predetermined temperature is discovered.

3. A method as recited in claim 1 in which the magnesium alloy to be cast is AXJ 530.

4. A method as recited in claim 1 in which the magnesium alloy to be cast is AXJ530 and the liquid environment to which the part may be exposed comprises salt water.

5. A method as recited in claim 1 in which the magnesium alloy to be cast comprises more aluminum than any other alloying element.

6. A method as recited in claim 1 in which the magnesium alloy to be cast comprises more aluminum than any other alloying element and the liquid environment to which the part may be exposed comprises salt water.

7. A method as recited in claim 1 in which the different sections of the part are in a cross-section of the part and include a surface and a central region of the part.

8. A method as recited in claim 1 in which the different sections of the part are in a cross-section of the part and include a surface and a central region of the part, and the liquid environment to which the part may be exposed comprises salt water.



This disclosure pertains to casting of magnesium alloys in metal molds having heating or cooling capability, or other casting processes that involve thermally managed dies or molds. More specifically this disclosure pertains to temperature control of a die casting process for a magnesium alloy article so as to improve the corrosion resistance of the die cast part.


There is a need to reduce the weight of automotive vehicles in order to conserve fuel. With this goal in mind low cost, creep-resistant magnesium alloys are being evaluated for cast components such as engine cylinder blocks, bed plates, oil pans, front engine covers, and other powertrain and vehicle components. Many such components may be produced by a high pressure die casting process in which structurally complex vehicle parts may be produced at a suitable production rate. Several magnesium-based alloys are available for consideration for the casting of such automotive vehicle components. These alloys typically contain at least ninety percent by weight magnesium. Many of the alloys contain aluminum (e.g., about two to nine percent) and manganese (often about 0.2-0.3 percent). Some of these die castable alloys also contain relatively small amounts of one or more of calcium, misch metal or individual rare earth elements, silicon, tin, strontium, and zinc. These magnesium alloys tend to have a microstructure comprising a principal phase of a solid solution of aluminum in magnesium and one or more other secondary phases depending on the composition of the alloy, the casting and solidification conditions, and any other thermomechanical processing of the cast part.

Whereas it has been industrial practice to generally optimize casting conditions for part quality and mechanical properties, this disclosure pertains to the new practice of thermal management of a die casting process (either high pressure or low pressure) to obtain such magnesium alloy castings with good resistance to galvanic corrosion arising from exposure of the vehicle part to salt and water containing environments.


The die casting of a magnesium alloy part requires the design of a multiple piece die (a mold), the sections of which are carried on platens of a die casting machine. When the machine is closed the die sections engage to form one or more cavities defining the shape of a part or parts to be cast. The die sections also contain flow channels for molten magnesium alloy to be injected into the cavity and for evacuation of air. The molten magnesium alloy is typically maintained at a suitable temperature depending on the composition of the alloy (for example, about 650° C. to about 700° C.). The molten alloy solidifies in the die cavity (ies) in a matter of a few seconds to minutes and the die sections are opened to permit removal by ejection of the cast part or parts.

A distinction may be made between high pressure and low pressure die casting. In high pressure die casting, molten metal is injected into the die cavity at extremely high speed, so high that the metal atomizes at the in-gates. The resulting ‘mist’ of metal permeates the die and condenses on all interior surfaces forming a dense, fine-grained skin. As this skin thickens, the metal solidification rate decreases and the inner regions of the casting form. This latter is referred to as the core. Casting process conditions such as die temperature, injection rate, and melt temperature affect both the composition and microstructure of the skin and core parts of the cross sections of the casting. In low pressure or gravity die casting, metal fill is much slower and there is generally less difference between the skin and core of the casting. Nevertheless, to the degree that die temperature affects the nature of the skin/core cross section, this invention may apply to any die casting process (high or low pressure) where thermal management of the die surfaces is practiced.

Provision is also made for temperature control of the casting dies. This may be accomplished by use of electrical resistance heating elements embedded in the dies or by the circulation of heated oil through passages machined in the dies. Temperature control may also be effected by external means when the die is open; such means including radiant or gas heating or cooling by means of sprayed liquids. The temperature of the dies has been set to achieve optimum mechanical properties of the part, timely solidification of the cast magnesium alloy parts without embrittlement or other impairment of the products, consistent with production objectives such as cycle time. There has been considerable latitude (for example, several hundred Celsius degrees) in the die face temperatures of die casting operations for producing the same part shapes with the same magnesium alloy and obtaining parts with useful physical properties. In brief, it is industrial practice for a particular casting geometry and metal to determine the processing parameters that are optimum for mechanical properties, reliability, and manufacturability. The present invention relates to predetermining a processing parameter that affects the corrosion behavior of a magnesium alloy.

Now, it is found that the die temperature may have a profound effect on the tendency of certain magnesium alloy parts to corrode, especially in a salt water environment. This corrosion has been attributed to a rate of solidification in the dies that produces significantly different microstructures between the surface (skin) and interior (core) of the die cast part. For example, an excessively high rate of solidification of a cross-section of a die cast part can result in a preponderance of one metallurgical phase being formed at the surface of a die cast part and a different phase of different composition formed internally. The proportions of different phases with different overall composition may form a galvanic couple between surface and core regions causing them to have higher susceptibility to accelerated corrosion in a salt water environment.

In accordance with an embodiment of this invention, casting trials at varying die temperatures may be made with a specific part and magnesium alloy composition to determine a die temperature (or temperature range) that yields a cast part that does not form the aforementioned “skin-core” two-phase microstructure in a cross-section that may render the part susceptible to unacceptable corrosion in, for example, a salt water corrosion test. Depending upon the size and shape of a die cast part and die temperature, solidification of the part may yield varying distributions of alloying elements across a cross-section of the magnesium-based part. Such variation may produce an electrochemical potential gradient between regions of the cross-section that may result in galvanic corrosion, especially in a wet salty environment.

For example, it may be desired to form a die cast part of magnesium alloy AXJ530. This alloy has a nominal composition, by weight, of about 5% aluminum, 3% calcium, 0.3% manganese, 0.15% strontium, and the balance magnesium, except for minor amounts of impurities such as traces of iron, nickel, and copper. As a melt of this composition solidifies some of the aluminum content is found in a primary magnesium-aluminum solid solution phase and the balance in a secondary magnesium-aluminum intermetallic compound phase (Mg17Al12). These phases have different compositions. In general, it is preferred that the two phases are inter-dispersed and distributed generally uniformly throughout a die casting section so as to avoid regions of different electrochemical potential in sections of the die cast part. A suitable operating temperature of the die is determined to produce a suitably uniform casting microstructure that is less conducive to such galvanic corrosion.

These and other objects and advantages of the invention will be apparent from illustrative specific embodiments.


An aim of this invention is to improve corrosion resistance of die cast magnesium alloys to enable their increased usage in automotive vehicle components in powertrain and body structures, especially in parts exposed to water and water- and salt-containing environments.

The development of low-cost, creep-resistant magnesium alloys will enable the use of magnesium to reduce the weight of powertrain cast components and other cast components. Magnesium alloys have been developed for such applications. For example, AXJ530, an aluminum (5%)—calcium (3%)—strontium (0.15%) magnesium alloy is very castable, typically has excellent corrosion resistance as measured by the GM9540P cyclic salt spray test, and has creep resistance approaching that of aluminum alloy 380.

In order to further assess development of this magnesium alloy composition, high-pressure die cast plates (five inches wide by seven inches long by 0.14 inch (3 mm) thick) were obtained from two different commercial die casting facilities. All specimens were prepared by high-pressure die casting [HPDC] using cold chamber HPDC machines and generally standard procedures. A first set of specimens (A) was cast on either a 420 ton Buhler Evolution or a 200 ton Frech, both located at a facility in Norway. Metal at 690° C. was hand ladled into the machines and cast into the dies, which were heated to 200° C. Klubertech HP1-415 die lubricant was applied to the die faces. A second set of specimens (B) was cast in a 700 ton Lester HPDC machine at a die casting facility in New York State. The melt temperature varied from 670 to 690° C. and the die lubricant was ChemTrend RDL9384. The die faces in the Lester were maintained between 350 and 400° C.

Sets of test specimens, one inch by two inches by three millimeters, were cut from the cast plates of both specimens A and specimens B. All specimens were polished with 1200-grit SiC paper and rinsed with deionized water and acetone prior to any testing.

The die cast specimens (A) experienced unexpected corrosion and corrosion rates in the ASTM B117 salt spray corrosion test. Corrosion rates were calculated from individual sample weight losses in the B117 test over a test period of ten days. The corrosion rates of the die cast A specimens were ten times the corrosion rates of the die cast B specimens. Chemical analysis of the die cast A and B specimens did not show differences in iron, copper, or nickel levels which have been found to affect corrosion behavior in magnesium alloys. This difference in salt spray corrosion testing was surprising because die castings of AXJ530 alloy have performed very well in various test programs which included static mechanical properties, tensile and compressive creep, and coolant corrosion tests. Die cast AXJ 530 parts have also performed very well in hot surface coolant corrosion (water/ethylene glycol coolants) and a galvanic corrosion test in four different experimental engine coolants. So the failure of the die cast A samples in the B117 salt spray test was surprising and initially unexplainable.


Weight Loss Measurements

The samples (both die cast A and B samples) were immersed for 1-3 days in 5 wt % NaCl solution with a pH of 6 at room temperature. After this period, the samples were removed from the solution, and immersed in 180 g/l of boiling chromic acid to remove the corrosion products. The samples were then rinsed with acetone, dried in air and weighed.

Hydrogen Evolution Method

A 1000 ml graduated glass beaker was filled with 5% NaCl solution and an inverted funnel was placed on the bottom. A 50 ml graduated burette was filled with test solution (5% NaCl) and inverted over the stem of the funnel, devoid of air bubbles. The test sample was slid under the funnel mouth and a record was made of time, temperature and pressure. The hydrogen gas evolved during the corrosion reaction (Mg+2H2O→Mg (OH)2+H2), was collected in the burette, and the amount of magnesium dissolved was calculated by the Ideal Gas Law. To check the effect of sample thickness on the rate of Mg dissolution, die cast A samples of varying thickness were mounted in epoxy resin (with surface masked), and only the cross sections were exposed to the salt solution. The exposed areas were measured for subsequent analyses. The calibration between weight loss and the gas collection method has been conducted with pure magnesium samples, and the accuracy is about ±10%.

Potentiodynamic Polarization Measurements

Polarization measurements were performed using die cast specimens A and B as working electrodes with a Solartron Model 1200 B Electrochemical interface and Frequency Response Analyzer in a flat cell configuration. A platinum grid served as the counter electrode, and an Ag/AgCl electrode in saturated KCl (+197 mV vs. standard hydrogen electrode) served as the reference electrode. 5% NaCl was used as the test electrolyte and the exposed working electrode area was 1 cm2. Open-circuit corrosion potential measurements were performed prior to all potentiodynamic polarizations tests. The polarization scan was started at about 250 mV cathodic to Ecorr (determined from the open-circuit corrosion potential measurements), and scanned at a rate of 1 mV/s in the noble direction to 1.2 V (vs. Ecorr).


Cross-sections of the tested samples were polished first with SiC paper up to 4000 grit and then with polishing cloth, rinsed with de-ionized water, and dried in air at room temperature. To reveal the fine microstructure, the polished samples were chemically etched using an etchant consisting of 50 ml of ethylene glycol, 1 ml of HNO3, 20 ml of glacial acetic acid, and 17 ml of distilled water. The cross-sections were then examined with a scanning electron microscope (Zeiss Model EVO 50).

Elemental analyses on the alloys were carried out by Inductively Coupled Plasma Atomic Emission Spectroscopy and Atomic Absorption Spectroscopy (ICP-AES/AAS). Finally, the alloys were analyzed with x-ray diffraction to identify any phase difference between the GM and Hydro samples.

Results and Discussion

The weight loss of die cast B specimens was about 8 mg/cm2 after an immersion of one day and remained constant for up to 4 days of immersion probably due to the formation of a protective passive film. The weight loss increases after the fourth day were possibly due to the rupture of this passive film. Since the sample was suspended in the solution with a Teflon string, both the cast plate surfaces and cross-sections were exposed to the corrosion medium. The corrosion was uniform for all the exposed surfaces, and the average corrosion rate was 2.8 mcd (mg/cm2/day).

The die cast A specimen 3 mm plate on the other hand showed a linear increase in weight loss with time, indicating that there was no retardation of corrosion with time. The average corrosion rate was 13 mcd; about 5 times greater than the die cast B specimens, which confirmed that the die cast B plates have a much higher corrosion rate.

Hydrogen evolution volume versus time tests for the die cast A and B specimens also demonstrated that the die cast A plates corroded faster (0.30 ml/cm2/hr) than the die cast B plates (0.08). Additional experiments were carried out to determine whether the cast plate thickness in the die cast A specimens had an effect on corrosion rate. Some A specimens were tested with all surfaces exposed (plate surface and cross-section) and some with only the cross-section exposed. The corrosion rate increased rapidly when the plate thickness increased from 1.27 to 3.5 mm when both the plate surface and cross-section were exposed, while only a slight increase in corrosion rate occurred when the plate thickness increased from 3.5 to 8.9 mm. However, when only the cross-sectional area was exposed, the corrosion rate increased continuously with plate thickness.

A photograph of a die cast A specimen when it was immersed in 5% NaCl solution clearly showed that in the A specimens, the surface was cathodic to the cross-section due to the presence of hydrogen bubbles on the plate surface. This was not observed in the die cast B specimens which showed uniform corrosion. Therefore, in the immersion test, when both the surface and cross-section areas are exposed, the surface layer (skin) remains intact while the cross-section corrodes. This mode of corrosion will induce a corrosion film that retards corrosion for a plate thickness greater than 3.5 mm, while when only the cross-section area is exposed, the corrosion product is less protective and the corrosion rates increase with plate thickness.

Corrosion potential vs. time plots were prepared for die cast A and B specimens. The tested specimens were mounted in epoxy and only the plate surface was exposed to the solution. The steady-state corrosion potential was very similar for samples from the two sources suggesting similar electrochemical behavior for the skin regions of these samples. The corrosion potential of the surface versus cross-section for the die cast B specimens showed that the steady-state corrosion potentials for these two regions were the same, and uniform corrosion behavior was also observed in immersion tests of the die cast B samples. However, with the die cast A specimens the corrosion potentials of the surface versus cross-section showed that the surface behaved cathodic (more noble corrosion potential) compared to the cross-section area, and galvanic corrosion would be expected when both surfaces are exposed.

An investigation into the casting conditions revealed that the die temperature for the die cast A specimens was 200° C. compared to 400° C. for the die cast B specimens. Based on this, it was believed that the lower die temperature resulted in a faster cooling rate leading to more Al becoming retained in solid solution in the Mg rich α phase at the surface compared to the core. On the other hand, the aluminum content of the Mg-rich phase was more uniform from surface to center for the die cast B castings done using a higher die temperature. Since the corrosion potential of the α phase becomes more negative (more active) with decreasing Al, the higher Al content at the surface (skin) would make it more cathodic compared to the core region. An examination of the surface microstructure of die cast A and B specimens revealed that, as expected, the average grain size for the B specimen is larger than the A specimen due to the higher die temperature used producing the die cast B specimen.

It is proposed to prepare for die casting of magnesium alloys as follows. The practice will be utilized for magnesium alloy parts that may be exposed to an identifiable corrosive environment such as contact with salt spray. And the practice may be particularly useful in die casting of aluminum-containing magnesium alloys which are used in high pressure die casting. Once a part is designed and dies made for die casting of the shape, tests will be conducted or modeled to establish a suitable die temperature for making the part that meets the production and other performance requirements. Having established a casting process window for the part, the specified die temperature or temperature range will be determined to yield die cast parts that have generally uniform metallurgical cross-sections that do not produce galvanically distinct regions that enable galvanic corrosion when the section is exposed to salty water or other aggressive environment.

The invention has been described in terms of illustrative embodiments that are not intended to limit the rightful scope of the invention.