About AFS and Metalcasting

Aluminum Metal Matrix Composites

This article was adapted from AFS Transactions papers 05-147 and 06-085 presented at the 109th and 110th Metalcasting Congresses. View 05-147. View paper 06-085.

The automotive industry recognizes that weight reduction and improved engine efficiency will make the greatest contribution to improved fuel economy with current powertrains. This is evidenced by the increased use of aluminum alloys in engine and chassis components. Aluminum and magnesium castings in this sector have grown in leaps and bounds over the past five years to help engineers design and manufacture more fuel efficient cars. 

The low density and high specific mechanical properties of aluminum metal matrix composites (MMC) make these alloys one of the most interesting material alternatives for the manufacture of lightweight parts for many types of vehicles. With wear resistance and strength equal to cast iron, 67% lower density and three times the thermal conductivity, aluminum MMC alloys are ideal materials for the manufacture of lightweight automotive and other commercial parts.

MMC’s desirable properties result from the presence of small, high strength ceramic particles, whiskers or fibers uniformly distributed throughout the aluminum alloy matrix. Aluminum MMC castings are economically competitive with iron and steel castings in many cases. However, the presence of these wear resistant particles significantly reduces the machinability of the alloys, making machining costs higher due mainly to increased tool wear. As a result, the application of cast MMCs to components requiring a large amount of secondary machining has been somewhat stifled.

Most components do not require the high performance capability of aluminum MMCs throughout their entirety. An un-reinforced cast alloy may accomodate the stresses in these areas. Reinforcement of only the high stress regions of a component is referred to as selective reinforcement. This approach to component design and manufacture optimizes the material for the application, reduces the cost of the cast MMC part and lowers machining costs.

Be Selective

Selective reinforcement has been applied to the production of ring grooves and combustion bowl rim regions of pistons for both gasoline and diesel engines and integral cylinder liners for production gasoline engines. For the ring groove and cylinder liner applications, the reinforcement predominantly improves wear resistance. Application of short fiber reinforcement to the bowl rim increases its elevated temperature properties. The use of selective reinforcement in a clutch disk and ventilated brake disk have been considered but not yet entered production.

Short fiber-reinforced composite materials provide higher performance benefits than the discontinuously reinforced particulate aluminum composites. Their manufacturing costs are slightly higher, but because it’s likely that the value of weight saved in a vehicle will increase drastically over the next few years, selectively reinforced cast aluminum components used for vehicle and engine weight reduction are increasingly probable.

The mechanical properties of selectively reinforced MMCs are a function of the fiber properties, the orientation of the fibers and the amount of fibers. Thus, the design engineer needs additional information, such as the mechanical properties of the cast aluminum MMC as a function of fiber quantity, chemistry and orientation. If the fibers are located in a planar, random orientation, the mechanical properties expectedly will be the same in all planar directions.

Fiber Content

Recent research results serve as a first step toward the design and production of selectively reinforced aluminum MMC components. In the study, vertical squeeze cast equipment was used to pressure infiltrate liquid AlSiCuMg alloy, a modified AA319.0, into rectangular preforms containing different volume fractions of alumina (NF and SF) and alumina-silica (CF) short ceramic fibers. The preforms were preheated in air to 1,360F (755C) and placed into a heated four-cavity steel die, which was sprayed with a die coating before each run.

After thermal treatment to either a T5 or T7 condition, four test blanks were sawn from each casting and radiographed to be classified by quality in accordance with ASTM E155. Test blanks with a quality level of Grade B or better were machined into tensile test specimens.

Tensile testing established the quantitative effect of reinforcement chemistry and quantity  and thermal treatment on mechanical properties.

As expected, the resultant tensile strength is a function of the amount of fiber added and the mechanical properties of the ceramic fibers. The tensile strength of the CF and SF fibers are of the same order of magnitude, and resultant tensile strength of the fiber-reinforced matrix (FRM) made with these fibers also is of the same order of magnitude. The higher tensile strength of the NF fibers is reflected by higher tensile strength in the aluminum FRM.   The elastic modulus of the FRM demonstrated a similar response. The T7 heat treatment increased both the yield and ultimate tensile strengths. Not expected were the higher elastic modulus values exhibited by the T7 heat-treated samples. Image analysis confirmed that fiber orientation was not the reason for the difference. Scanning electron microscopy and x-ray diffraction analysis of T7 heat treated samples confirmed the presence of an aluminum-magnesium oxide (spinel) at the interface between the fibers and the aluminum matrix.

The chemical reaction at the interface could have been the cause of the increased strength of the fiber-matrix interface and provided enhanced load transfer from the matrix to the fibers. The result was the increased elastic modulus observed for the T7 heat-treated materials.

Optical metallography showed that the short fibers have essentially a planar random orientation in the FRM. However, as the volume percent of fiber in the preform increases, there is a slight preferential orientation in the tensile axis of the specimens. The data also showed that fiber diameter influenced preferred orientation. The larger diameter NF fibers orient differently than the smaller diameter CF and SF fibers.

Costing MMCs

The property testing results clearly indicate that the highest performance is obtained by reinforcing the 319 metal matrix with the NF chopped fibers. However, knowing the cost to obtain this performance is critical for future applications. In this study, the collected mechanical property data and estimated costs to manufacture generic short-fiber preforms were used to calculate the relative values (cost per property) of the three fibers used to make the FRM test castings. These calculated values provide a guideline for selection of fiber type to meet the needs of the application.

Because preform size has an impact on the ratio of fiber cost to processing cost, the relative costs to manufacture short-fiber preforms were estimated for small, medium and large preforms. The volume of a small preform is 1.5 cu. in. (25 cu. cm); a medium preform is 4.8 cu. in. (75 cu. cm); and a large preform is 9.2 cu. in. (150 cu. cm).

Fiber Values

The relative cost values for the preforms were divided by the relative property values to obtain a relative value for each type of fiber. These are relative fiber values (cost per performance) and are not to be used for estimating actual costs to fabricate preforms for specific applications.

Preform size has a significant effect on relative fiber value. Small preforms showed little difference in relative fiber value, making it difficult to prove that one fiber has higher value than the others. This parity among fibers suggests that the use of small preforms could be more economically beneficial. The NF chopped fibers provide higher properties at lower volume fractions, but according to the value calculations, they may not be cost effective in preforms larger than 1.5 cu. in. (25 cu. cm). METAL

This article was adapted from AFS Transactions papers 05-147 and 06-085 presented at the 109th and 110th Metalcasting Congresses. View paper 05-047. View paper 06-085.