Metal Matrix Composites

Metal Matrix Composites

P1: GBK/MBQ P2: FYD/MAG QC: FYD Final Pages Encyclopedia of Physical Science and Technology Qu: 00, 00, 00, 00 EN009D-424 July 19, 2001 18:55 ...

352KB Sizes 1 Downloads 61 Views

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

Qu: 00, 00, 00, 00

EN009D-424

July 19, 2001

18:55

Metal Matrix Composites Franck A. Girot Azar P. Majidi Tsu Wei Chou University of Delaware

I. II. III. IV. V.

Components Manufacturing Process Fiber–Matrix Interface Mechanical Properties Structure–Performance Maps

GLOSSARY Rheocasting Casting technique in which vigorous agitation of a semisolid alloy (temperature is between liquidus and solidus) before casting prevents the formation of dendritic properties in the primary phase. Squeeze casting Casting technique in which liquid metal is pressed into a ceramic preform. This process provides close tolerance dimensions and high integrity. Whiskers Short, single-crystalline fibers with the axis of growth in a crystallographic direction. They are 0.1– 1 µm in diameter and up to several hundred micrometers long.

THE LARGE AMOUNT of research done on metals (e.g., aluminum, magnesium, copper, lithium, titanium, nickel, and chromium) since the 1970s has resulted in the development of materials with high specific properties, good thermal or electrical conductivities, and good corrosion resistance at low production costs. However, the field of application of these materials is restricted by their

low stiffness, the rapid drop of their properties at high temperature, and their poor fatigue resistance. It has been well established in the past several years that the incorporation of ceramic fibers or whiskers into metal alloys is a very efficient way to improve their mechanical properties, particularly at high temperatures (above 300◦ C). Furthermore, these metal matrix composites (MMC) present different advantages with respect to organic matrix composites (where the matrix is a thermoset or a thermoplastic resin): they have higher transverse properties and can be used at elevated temperature. They also show better rupture energies, good thermal or electrical conductivities, good gas tightness, and insensitivity to water or ultraviolet ray emission. The field of application of metal matrix composites has extended to aerospace, automotive, and electronic industries. The number of these applications is increasing as better processing conditions give rise to cheaper and higher performance materials. The purpose of this contribution is first to describe the different components used in the manufacturing of metal matrix composites, then to briefly review the most widely

485

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

486

Metal Matrix Composites

used manufacturing techniques now available, and finally to present the different properties we can expect from these materials.

I. COMPONENTS The resulting properties of MMCs depend on (1) the ability of the matrix to support the fibers and to transfer the applied load to them and (2) the capacity of the reinforcement to sustain very high loads. The improvement of the properties of fibers and matrices is a main factor although the control of the interface and the choice of the different components are also important. A. Matrices The nature of the matrix will depend on the potential application of the composite. In that respect, one will choose aluminum or magnesium alloys for applications where low densities and high specific properties are required; nickel, titanium, or chromium for high-temperature applications; copper or aluminum when good electrical conductivities are necessary; copper or zinc when a low friction coefficient is important; and lead or lithium for batteries. Before the fabrication of the composite, several difficulties will have to be overcome: 1. The nonwetting of most of the ceramic fibers by liquid metals and alloys can be solved by coating the fiber surface with TiB2 for aluminum-, lead-, or copper-based alloys; K2 ZrF6 for aluminum alloys; and Ni or Cu and SiO2 for aluminum- or magnesium-based alloys. Alloying elements such as lithium or magnesium can also be added to the pure metal to improve the wetting. 2. The chemical reactions between ceramic fibers and the matrix can be minimized by coating the fibers with a protective layer, usually a ceramic material, by adding alloying elements to the matrix to obtain a system in equilibrium, or by reducing the contact time between fibers and liquid alloys. 3. The damage caused to the fibers can be minimized by optimizing the processing parameters.

FIGURE 1 Fiber properties 1.

of elements such as boron, carbon, nitrogen, oxygen, aluminum, or silicon. The different types of reinforcements available are in the forms of particles, fibers, and whiskers. The properties of various reinforcements are compared in Figs. 1 and 2. The potential applications of MMCs reinforced with particles are very limited, so only the last two types of reinforcements are presented. 1. Fibers Resulting from a Solid or Liquid Precursor Carbon, boron nitride, silicon carbide, and alumina fibers have been obtained in this manner. The process consists of five steps: 1. Preparing and melting an organic precursor: polyacrylonitrile (PAN), usually for carbon fibers; boric anhydride for BN fibers; polycarbosilane for SiC fibers; or formacetate with alumina particles for alumina fibers. 2. The precursor is drawn out from the furnace (with viscosity around 1000 P) and reduced to a diameter of several micrometers.

B. Reinforcements Because the fibers have to support high loads, they must have a high Young’s modulus E and ultimate tensile strength (UTS). They must also have good fatigue properties, a low density, and a high-temperature capability. The only solids meeting these requirements are made of light elements with strong atomic bonds. So, it is not surprising that the different fibers available on the market are made

FIGURE 2 Fiber properties 2.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

487

Metal Matrix Composites

3. A stabilization treatment is then performed in order to make the fiber nonfusible (180–450◦ C depending on the fiber nature). 4. The fiber is pyrolized at a high temperature (1000– 1400◦ C). 5. A very high temperature treatment (graphitization) follows the pyrolysis in the case of carbon (3000◦ C) and boron nitride (1800–2000◦ C). After manufacturing the fibers are given a sizing treatment in order to protect them from damages and abrasion and to improve their compatibility with the different matrices they have to reinforce. Two types of carbon fibers are currently available: highmodulus (HM) fibers with a modulus of ∼340–700 GPa and UTS of ∼1700–2400 MPa and high-strength (HS) fibers with a modulus of ∼235 GPa and UTS of ∼300– 4500 MPa. These fibers can be easily woven into various shapes, but are very sensitive to oxidation and cannot be employed in oxidizing atmospheres above 600◦ C. In inert atmosphere they maintain their properties up to 1500◦ C. Boron nitride fibers are of interest because of their electrical and thermal properties as well as their chemical stability at high temperatures. They can be used in air up to 800–1000◦ C. Alumina and silicon carbide fibers have lower mechanical properties than carbon fibers but better chemical stability up to 1000–1200◦ C. They can be woven as carbon fibers but with some difficulties in the case of alumina fibers. 2. Fibers Resulting from a Gaseous Precursor These types of reinforcement are essentially monofilaments of large diameter (100–200 µm) made of boron, boron carbide, or silicon carbide. A fibrous substrate (a tungstem wire of 12 µm in diameter or a carbon wire of 30 µm) is run through a chemical vapor deposition (CVD) reactor, where gaseous reactants are introduced (hydrogen and boron trichloride or silane). The wire substrate is resistance heated to 1100– 1400◦ C. Hydrogen and gaseous precursors then react in the reactor leading to the deposition of solid boron, boron carbide, or silicon carbide on the wire surface. After repeated depositions the filament reaches its final diameter and is then wound for use. The mechanical properties of these filaments are comparable to those of carbon fibers (E ≈ 400 GPa; UTS ≈ 3000–4000 MPa). 3. Whiskers Whiskers are short single-crystalline fibers (0.1–1 µm in diameter and up to several hundred micrometers in

length) with the axis of growth in a crystallographic direction. Their manufacturing process is based on usual processes of crystallogenesis with different mechanisms such as mass transport in gaseous phase, germination, and growth. Silicon carbide, silicon nitride, and alumina whiskers are manufactured by this process but at high costs. Recently, a process that uses rice hules has been developed, which allows the manufacturing of silicon carbide whiskers at low price. It is well known that pyrolysis residues of some plants (e.g., rice hules) have a high silica content. These secondary products of the cereal industry are inexpensive and available in large quantities. The pyrolysis of rice hules leads to a mixture of organic compounds and silica. This mixture is then heated to high temperature, leading to the coking of the organic compounds, which transform into carbon. This carbon reduces the silica to give silicon carbide. Whiskers are characterized by an elastic behavior up to rupture, and an ultimate tensile strength higher than that of continuous fibers, reaching the theoretical value (UTS ≈ E/10).

II. MANUFACTURING PROCESS The different techniques that are presented here are adapted for mass production (particularly the liquid- and semisolid-phase manufacturing processes) or for applications where a high purity of the different components is required (spray or powder metallurgy processes for electronics or aerospace). A. Liquid-Phase Manufacturing Process Liquid-phase fabrication methods are particularly suited for producing composite parts of complex shapes and the apparatus needed is relatively simple. Uniform fiber distributions can usually be achieved with little porosity in the matrix material. Several impregnation techniques have been used. 1. Vacuum Impregnation Elements such as calcium or magnesium have been used to improve the wetting. Lithium is the most efficient element for the impregnation of alumina fibers. The alumina preforms are impregnated by aluminum–lithium (3%) alloy under vacuum in order to avoid the oxidation of lithium and to facilitate the infiltration by the liquid metal. Optimization of the process depends on the optimization of the following parameters: lithium content, preform temperature, and contact time between fiber and liquid metal.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

488

Metal Matrix Composites

3. Squeeze Casting

FIGURE 3 Vacuum impregnation.

The mechanical properties of these materials are generally high and close to the predicted values (Fig. 3). 2. Fluorides Treatment A new casting process for the impregnation of fibrous ceramic preforms by aluminum alloys has been proposed in order to provide a low-cost technique for mass production of aluminum matrix composites. The method is based on a wetting enhancement treatment that involves the deposition of a fluoride K2 ZrF6 from an aqueous solution at 90–100◦ C, within the pore network of the preform. The chemical reaction that occurs between liquid aluminum and the fluoride allows the dissolution of the alumina scale on the liquid-aluminum alloys. The alumina scale tends to prevent good wetting of most ceramic fibers. After such a treatment, a ceramic porous preform can be merely impregnated by gravity in a permanent mold heated at ∼600◦ C, only by capillary effect (Fig. 4). Although the process is easily performed, accurate experimental conditions (e.g., fluoride quantity and preform temperature) are necessary for a proper impregnation and good fiber–matrix bonding. Other types of fiber coatings (e.g., Ni, Si3 N4 , and SiC) can be used along with the fluoride treatment to further control the interface properties.

Squeeze casting is a technique of liquid metal working under pressure that allows the manufacture of near net shapes with close tolerance dimensions and a high integrity. Usually, a preheated ceramic porous preform is placed in a die cavity, and a liquid metal is poured on. A piston applies the pressure on the liquid metal forcing it to impregnate the preform, as illustrated in Fig. 5. Two techniques can be distinguished: (1) direct squeeze casting (previously described) and (2) indirect squeeze casting where pressure enables to feed several molds. Squeeze casting offers two major advantages: 1. Pressure compensates for the poor wetting of the ceramic preform with metal alloys (e.g., aluminum, copper, and zinc) regardless of the size of the reinforcement and the residual porosity of the preform before impregnation. 2. Solidification under pressure eliminates macro- and micropores leading to a better metal lurgical quality of the composite. The depths of impregnation depend on numerous parameters among which the most important are the preform and liquid-alloy temperatures, the physical characteristics and volume fraction of reinforcement, the contact angle between the ceramic and the liquid alloy, and obviously the pressure as well as the impregnation speed. Actually, this process seems to be the most efficient because it allows the manufacture of finished parts at low cost and in a large scale. So, it is particularly adapted to the automotive industry for the manufacturing of connecting rods or pistons. B. Semisolid-Phase Manufacturing Process: Compocasting The compocasting process is the adaptation of the rheocasting technique to the manufacturing of discontinuous reinforced MMC. Rheocasting is a technique developed

FIGURE 4 Fluoride treatment for wetting enhancement.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

489

Metal Matrix Composites

FIGURE 5 Squeeze casting is shown in steps (a)–(d).

by M. C. Flemings, which requires the vigorous agitation of a semisolid alloy (the temperature of the alloy is between the liquidus and the solidus) before casting so that the primary phase is nondendritic, giving a slurry with thixotropic properties and leading to a globular microstructure in the solidified alloy (Fig. 6). MMC can be processed according to the following sequences: 1. The liquid-metal alloy is rapidly cooled down to temperatures in the liquidus–solidus range and is followed by a vigorous mixing, which prevents the primary phase to form dendrites and giving the mixture a low apparent viscosity. 2. At this steady-state temperature, the reinforcement is added to the melt and dispersed within the liquid by the solid primary phase.

3. Then the compound is cast when its fluidity is sufficient or a complementary technique such as injection or squeeze casting is used to form the composite when the viscosity is too high. Usually, the resulting composites have a porosity content high enough to decrease significantly the mechanical properties. The complementary technique often allows to reduce the porosity content and to improve the metallurgical quality of the composite. However, it is possible to obtain cast composite parts with a lowporosity content by performing the mixing under vacuum. Two variations of this process, namely mixing the reinforcement in a liquid matrix and reheating the comopund to melt the matrix before casting, can overcome some of the drawbacks of this method but result in a more nonuniform fiber distribution.

FIGURE 6 Continuous compocaster.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

490

Metal Matrix Composites

The compocasting process is particularly attractive for automotive applications (connecting rods), for friction application (bearings), and for electronics (heat sinks and alumina chip carrier supports). C. Solid-Phase Manufacturing Process 1. Spraying and Diffusion Bonding The spraying device has a hot zone produced by a gas flame, a gas explosion, or an electric arc through which a gas flows. Metal particles are fed into the gas flow, melted in the hot zone, and sprayed onto an assembly of fiber arrays. In the electric arc process, the arc is used to produce molten particles from the wire electrode, and the gas flow sprays them onto the fiber surface. A thin film of matrix is then formed, which binds to the fibers without physical or chemical damage to the fibers. Diffusion bonding is then used to obtain highdensity composites. Preformed sheets are laid in predetermined orientations to achieve the fiber volume fraction and thickness required. Press forming achieves the bonding between the different sheets of fibers and matrix through the application of pressure and temperature. Diffusion bonding can also be performed on alternative stacking of metal sheets and fiber weaves. 2. Powder Metallurgy The incorporation of the reinforcement can be performed in the solid state by blending a powder of metal alloy and short fibers, whiskers, or particles. The reinforcing elements, whose volume fraction can be rather high, are generally well distributed. Pressure is then applied in order to form the mixture in near net shape part and to densify it before sintering. The part is heated to a high temperature, which depends on the nature of the metal alloy, and maintained at that temperature during sintering. Bonding is achieved usually by a diffusion phenomenon leading to composite with a good metallurgical quality. However, the uniaxial or isostatic pressures, which are applied before sintering, are quite high so that some degradation of the reinforcement occures, particularly when this sequence takes place at room temperature. This mechanical degradation can be reduced by applying pressure at high temperatures, eventually within the liquidus–solidus range. A schematic representation of MMC powder processing is given in Fig. 7. This process is less suitable when the reinforcement is long, but is particularly interesting in the case of reinforcement of small dimension such as whiskers. Moreover, secondary work can be applied to whisker-reinforced MMC by using processes such as hot rolling, extrusion, and drawing.

FIGURE 7 Metal matrix composites by powder metallurgy.

III. FIBER–MATRIX INTERFACE Metal matrix composites are generally nonequilibrium systems. Therefore, during fabrication or subsequent hightemperature exposure, diffusion of elements across the interface takes place, which often causes severe degradation of the fiber strength. Degradation is due to one or more of the following mechanisms; (1) formation of a brittle interfacial reaction layer; (2) reduction in fiber diameter; and (3) formation of voids in the fiber due to the rapid, outward diffusion of the elements. A thin reaction layer is often desirable as it promotes load transfer between the fiber and the matrix and thus increases the strenght of the composite. A thick layer, on the other hand, is detrimental to the fiber strength because it fractures at a low stress and triggers fiber fracture. A fracture mechanics model for the strength of fibers with brittle reaction zones has been suggested. A summary of the interfacial characteristics of various metal matrix composites is given in the following. A. Aluminum Matrix Composites A severe reaction occurs between carbon fibers and aluminum alloys at temperatures above 550◦ C. Complete disappearance of the fibers may even take place at sufficiently high temperatures or after very long exposures. The primary reaction product is the brittle aluminum carbide, Al4 C3 , phase which is often blocky, lath-shaped, or cuboidal. It has been found that pitch-based fibers are less prone to degradation than PAN-based fibers. Interfacial reactions for alumina FP and SiC fibers in aluminum matrices are less severe. The reaction products in FP/Al–Li composites are predominantly LiAl5 O8 spinel and LiAlO2 and form at the grain boundaries in the fiber. Lithium is often added to the aluminum matrix to facilitate the wetting of the alumina fiber. For the system FP/Al–Cu (4.5%), research has shown no significant reaction.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

491

Metal Matrix Composites

The Nicalon SiC fiber is readily attacked by the molten aluminum to form Al4 C3 by the following equation: 4Al + 3SiC → Al4 C3 + 3Si Although it may appear from this reaction that the presence of Si in the aluminum matrix will suppress the reaction, experiments with Al–Si (14.5%) alloy have shown that the opposite is true and the reaction zone actually thickens in the presence of Si in the matrix. The increased reaction in the Si-containing matrix is due to the presence of SiO2 and free carbon in the Nicalon SiC fiber. In the case of CVD SiC monofilaments or some SiC whiskers where the reinforcement or some SiC whiskers where the reinforcement exhibits a higher degree of purity, the addition of silicon to aluminum may show the expected advantage. The beneficial effect of Si alloying on the stability of pure SiC–Al system has been demonstrated. In the case of SiC-whisker-reinforced aluminum-alloy composites, little reaction has been reported. Boron filaments also react heavily with aluminum and aluminum alloys to form brittle interfacial zones, which severely lower the strenght of the composite. The reaction layer is primarily aluminum diboride, AlB2 , but may also contain other phases depending on the composition of the aluminum alloy. For example, in the case of B/6061 Al composites, a layer of AlB12 also forms on the fiber side in addition to the AlB2 layer. A fiber fracture model for the thermally induced strenght degradation of B–Al composites has been proposed. The model assumes a parabolic growth rate for the brittle interfacial reaction layer. B. Magnesium Matrix Composites Although magnesium alloys have a better wettability than aluminum alloys, there are still cases where fiber wetting is a limiting factor in the fabrication of magnesium-matrix composites. Katzman has recommended a silicon dioxide, SiO2 , fiber coating from an organometallic precursor solution to promote fiber wetting by magnesium alloys. The coating was specifically developed for carbon fibers, but has also been successfully used to fabricate magnesium composites with FP alumina or Nicalon SiC fibers. In the case of high-modulus pitch-based carbon fibers, first a layer of amorphous carbon is deposited on the surface of the fiber to facilitate the adhesion of SiO2 to the fiber. In cast composites, the SiO2 -coated carbon fiber reacts with the magnesium matrix to form an interfacial layer consisting of MgO and Mg2 Si. MgO is also the interface reaction product in FP-alumina-fiber-reinforced magnesium composites. Unlike the reaction layers in aluminum matrix composites, however, the presence of the MgO

reaction layer in magnesium matrix composites does not appear to seriously degrade the fiber strength. No reaction has been observed in diffusion-bonded carbon fiber–magnesium composites. C. Titanium Matrix Composites Titanium and its alloys are highly reactive and attack most fibers. In boron-fiber-reinforced titanium composites, TiB2 and TiB, reaction layers readily occur at the interface in the form of concentric rings with the TiB2 ring being next to the fiber. Boron atom is smaller than titanium atom. Therefore, the outward diffusion of boron is faster than the inward diffusion of titanium, which causes void formation near the tungsten core in the filament. B4 C coating of the fiber provides a diffusion barrier which slows down the reaction. A SiC coating is much less effective as it is quickly destroyed due to the formation of titanium carbide and silicides. After longer annealing times Si diffuses into the matrix and Ti into the fiber, thus causing the formation of TiB2 . Titanium carbides and silicides are also the primary reaction products in composites of CVD-produced SiC monofilaments in Ti and Ti–6Al–4V matrices. The reaction consumes both the fiber and the matrix and, therefore, causes a severe reduction in the fiber diameter.

IV. MECHANICAL PROPERTIES Fiber reinforcement significantly improves the strength and stiffness; resistance to fatigue, friction, and wear; and creep of the pure metals and alloys. The above improvements, however, are achieved at the cost of ductility and toughness, which result from the severe retardation of the plastic deformation in the matrix by the fibers. The strength and stiffness in tension, compression, and shear of MMCs and the effect of thermal exposure on these properties have been extensively studied for aluminum, magnesium, and titanium matrix composites. These properties are often considerably below the predictions of the rule of mixtures. The reason is primarily fiber degradation during fabrication or subsequent thermal exposure as described above. The effects of fiber orientation and fiber spacing and diameter on the strength and stiffness of metal matrix composites have been reported. J. L. Christian studied the fatigue strength of a variety of MMCs and found that the fatigue resistance of unidirectional Al matrix composites is higher in the 0◦ direction but lower in the 90◦ direction than those of Ti matrix composites. Heat treatment significantly improved the transverse fatigue strength but had little or

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

492 no effect on the axial fatigue resistance. In Ti–6Al–4V matrix reinforced with B, B(B4 C), and SiC filaments, J. M. Quenisset and coworkers found that a short thermal exposure at 850◦ C increased the resistance to fatigue crack propagation of composites reinforced with B and B(B4 C) fibers, while long exposures caused a severe drop in the strength and fatigue resistance. In SiC fiber composites, on the other hand, a short thermal exposure strongly decreased the fatigue resistance; however, the effect slowed down with time until, after 10 hr of exposure, all three composites exhibited a more or less similar degree of degradation. Several other investigations have also been conducted on the fatigue characteristics of MMCs under both mechanical and thermal cycling loadings. While the continuous fiber MMCs have shown a considerably higher axial fatigue resistance than their matrices, a reduction in the fatigue resistance has been reported in a randomly oriented short Saffil δ-aluminafiber-reinforced aluminum composites. Fiber fracture and a strong fiber/matrix bond were cited as possible reasons for the reduction. The friction and wear properties of MMCs can be substantially better than those of the unreinforced matrix material. The wear rate has been found to be a strong function of the fiber orientation. The lowest rate is achieved when fibers are normal to the sliding surface. In carbonfiberreinforced metal matrix composites, the type of the fiber has a significant effect on the wear rate. The lowmodulus PAN-based fibers give the lowest wear and friction rates, while the high-modulus PAN-, rayon-, and pitch-based fibers give significantly higher rates of wear and friction. The greatest benefit of fiber reinforcement of metals is to reduce their creep rate, thus increasing their maximum use temperature. Mechanistic models of creep of MMCs have been developed by A. Kelly and K. N. Street, M. McLean, and M. Lilholt. These models treat both cases of rigid fibers in a creeping matrix, characterized by ceramic fibers such as SiC or Al2 O3 in metallic matrix, and creeping fibers in a creeping matrix, represented by MMCs reinforced by refractory metal wires. In a composite containing perfectly aligned, continuous fibers, the creep rate progressively falls until it reaches zero at a certain strain level. Short-fiber composites and continuous-fiber composites under off-axis loading, however, both exhibit a steady-state creep rate. In addition to the composites based on unidirectional lamina or short fibers, multiaxially reinforced textile MMCs have also been developed and characterized. A comprehensive investigation of the strength, toughness, impact resistance, and friction and wear properties of three-dimensional braided FP alumina-fiber-reinforced

Metal Matrix Composites

FIGURE 8 Structure–performance map 1.

Al–Li composites has been conducted by A. P. Majidi and coworkers.

V. STRUCTURE–PERFORMANCE MAPS Considerable effort has been devoted by researchers to study the thermoelastic behavior of metal matrix composites and to evaluate the effectiveness of various reinforcement concepts. However, the analysis and experimentation performed on advanced composites are usually reported for individual systems; it is thus difficult

FIGURE 9 Structure–performance map 2.

P1: GBK/MBQ

P2: FYD/MAG

QC: FYD Final Pages

Encyclopedia of Physical Science and Technology

EN009D-424

July 19, 2001

18:55

493

Metal Matrix Composites

SEE ALSO THE FOLLOWING ARTICLES ALUMINUM • CERAMICS • COMPOSITE MATERIALS • FRACTURE AND FATIGUE • METALLURGY, MECHANICAL • POLYMERS, MECHANICAL BEHAVIOR

BIBLIOGRAPHY

FIGURE 10 Structure–performance map 3.

to acquire a more comprehensive view. In this section, the results of extensive studies in the modeling of thermoelastic behavior of unidirectional laminated composites, as well as two-dimensional and three-dimensional textile structural composites, are integrated in the form of structure–performance maps. The geometric configurations of the fiber reinforcements consist of angle-ply (+θ/−θ) and crossply (0◦ /90◦ ) laminates of unidirectional laminae, two-dimensional fabric materials with increasing fabric repeating pattern size ranging from plain weave to eight-harness satin weave, and three-dimensional braided preforms. The matrix metals included in this study are aluminum, titanium, magnesium, and a superalloy (MAR-M-200 of Martin Marietta Co.). The fiber materials are B, SiC, Al2 O3 , and refractory wires of tungsten alloy (W-2 pct ThO2 ) and molybdenum alloy (TZM). Figures 8–10 show the performance of the composite systems in terms of the Young’s modulus (E x and E y in the two mutually orthogonal directions), shear modulus (G x y ), and thermal expansion coefficient (αx and α y in the two mutually orthogonal directions). In the braided composites the braiding angle, θ, is defined as the angle between the braiding axis and the average orientation of the fiber segments in the preform. Through the construction of structure–performance maps, the relative effectiveness and uniqueness of various reinforcement concepts can be assessed. These maps should guide engineers in material selection for structural design and researchers in identifying the needs for future work.

Chou, T. W., and Kelly, A. (1985). Composites 16, 187–206. Chou, T. W., and Yang, J. M. (1986). Metall. Trans. A 17, 1547. Clyne, T. W., and Withers, P. J. (1995). “An Introduction to Metal Matrix Composites,” Cambridge Univ. Press, Cambridge, MA. Clyne, T. W., and Simancik, F. (Eds.). (2000). “EUROMAT 99, Vol. 5, Metal Matrix Composites and Metallic Foams,” Wiley, New York. Delmonte, J. (1981). “Technology of Carbon and Graphite Fiber Composites,” pp. 41–87. Van Nostrand–Reinhold, Princeton, NJ. Girot, F. A., Fedou, R., Quenisset, J. M., and Naslain, R. (1987). “Proceedings of the American Society for Composites, Second Technical Conference,” Technomic, Lancaster, PA. Hayashi, T. (Ed.) (1982). “Progress in Science and Engineering of Composites,” Proc. ICCM 4, Tokyo, pp. 113–120, 1297–1305, 1315–1322, 1465–1472, Japan Soc. Composite Materials, Tokyo. Karandikar, P. G., and Chou, T. W. (1990). Manufacturing of C/Al composites by gravity infiltration and their characterization. In “Fundamental Relationships Between Microstructure and Mechanical Properties of Metal Matrix Composites,” (M. N. Gungor et al., Eds.), pp. 59–69, TMS, Warrendale, PA. Karandikar, P. G., and Chou, T. W. (1991). Characterization of aluminummatrix composites made by compocasting and its variations. J. Mater. Sci. Majidi, A. P., and Chou, T. W. (1987). Structure-reliability studies of three-dimensionally braided metal matrix composite. In “Proceedings of Sixth International and Second European Conference on Composite Materials,” Elsevier Applied Science, London. Majidi, A. P., Yang, J. M., Pipes, R. B., and Chou, T. W. (1985). Mechanical behavior of three-dimensionally woven fiber composites. In “Proceedings of the Fifth International Conference on Composite Materials, ICCM-IV” (W. C. Harrigan, J. Strife, and A. K. Dhingra, Eds.). The Metallurgical Society of AIME, Warrendale, PA. Martineau, P., Pailler, R., Lahaye, M., and Naslain, R. (1984). J. Mat. Sci. 19, 2749–2770. Mileiko, S. T. (1997). “Metal and Ceramic Based Composites,” Elsevier, New York. Shehata, M. T., Leduc, T. R., Lemay, I., and Louthan, M. R. (1995). “Metallographic Techniques and Characterization of Composites, Stainless Steels, and Other Engineering Materials,” ASM International, Washington, DC. Shetty, H. R., and Chou, T. W. (1985). Metall. Trans. A 16, 853–864. Terry, B., and Jones, G. (1990). “Metal Matrix Composites: Current Developments and Future Trends in Industrial Research Applications,” Elsevier, New York. Viala, J. C., Bosselet, F., Fortier, P., and Bouix, P. (1987). Chemical interaction between silicon carbide nicalon fibers and liquid aluminum or aluminum-silicon alloys. In “Proceedings of the Sixth International and Second European Conference on Composite Materials, ICCM IV & ECC 2” (F. L. Mathews, N. C. R. Buskell, J. M. Hodgkinson, and J. Morton, Eds.), Vol. 2, Elsevier Applied Science, London. Voyiadjis, G. Z., and Kattan, P. I. (1999). “Advances in Damage Mechanics: Metals and Metal Matrix Composites,” Elsevier, New York.