Metal Matrix Composites

Chapter 4

Metal Matrix Composites

. . . . . . . . . . . . . . . . * * * * * * * * * * * n ” * ’ ”

Current Applications and Market Opportunities . . . . . . . .Longer Term Applications.. . . . . ● . . . . . . . . . . . . . . . . . . .

Research and Development Priorities . . . . . . . . . . . . . . . .Introduction ● ● ● ● ● ● ● ● ● . ● . ● ● ● ● ● ● . ● ● ● ● ● ● ● ● ● . ● ● ● .-. .

Properties of Metal Matrix Composites ● ● ● ● . ● ● . . ● . . . ● ● .Discontinuous Reinforcement ● . ● . . . . . . . . . . . . . . . .’. . ..

. Continuous Reinforcement ● . . ... .. . . . . . . . . . . . . . . .MMC Properties Compared to Other Structural Materials

●Design, Processing, and Testing . . . . . . . . . . .. Design . . . ● . ● ● ● . . . . . . . ● ● * ● ● * ● ● ● . .

Processing. ● . . * . . . ● ● . . . . . . ● ● . ● ● ● . . . . ●

Costs. ● . . . ● ● .. ● , ● . . . . . . . . . . . ● ● ● ● ● ● . .. Testing . ● ● . . . . . ● ● . . . . ● . . . . . . . ● ● ● ● ● ●

Health and Safety . . . . . . . . . . . . . .. . . . . . . Applications.. . . . . . . . . . . . . . . . ● . . . . . . . . .

Current Applications .. . . . . . . . . .......Future Applications . .............. ...Markets . ● . . . ● . . . ● . . . .

Research and Development Priories . . . . . . .Very Important.. . .............. . . . . . .

Important . . . . . . . . . . . . . . . . . . . . . . . . . . . .Desirable . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Tables

,....., .*...**.... ..*.. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . . . . . .. . . . . . ,.,... . . . . . .*.*.. . . . . . . .. . . . . . . . . . .● ☛✎✎☛✎✎ ✎ ✎ ✎ ✎

✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎

✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎ ✎

Page..... . . . . . . . . . . 99. . . . . . . .... . . 99. . . . . . .... ● .. 99. . . . . . . . . . . . . ... 99. . . . . . . . . . . . . . . . . 100. . . . . . . . . . . . . . . . . 100. . . . . . . . . . . . . . . . . 100. . . . . . . . . . . . . . . . . 102. . . . . . . . . . . . . . . . . 102...... . . . . . . . . . 106... ●☛✎✎✎✎✎ . . 107. . . . . . ...,.. ● . . . 107. . . . . . . . . . . . . . . . . 110. . . . . . . . . . . . . . . . . 111.... ● . . . . . . . . . 111. . . . . . . . . . . . . . . . . 112. . . . . . . ...... . . 112. . . . . . . . . . . . . . . . . 113.... . . . . . . . . . . 115.... . . . . . . . . . . 115

● 115. . . . . . . . . . . . . . . . .. . . . . . . ...... . . 116. . . . . . . . . . . . . . . . . 117

Table No.4-1. Costs of a Representative Sample of MMC Reinforcements. . ............1014-2. Structural Properties of Representative MMCs, Compared to Other

Materials . . . . . . . . .. .. ... . . . . . . . . . . .............................1034-3. Strength and Stiffness of Some Fiber-Reinforced MMCs. . . . . . . . . . . .. ....104

4-4. Properties of 6061 Aluminum Reinforced With Silicon Carbide Particulate .1054-5. MMC Manufacturing Methods .. . * . . . . . . . . . . . . . ...................1084-6. Selected MMC Processing Techniques and Their Characteristics . ........109

i

Chapter 4

Metal Matrix Composites

FINDINGS

Metal matrix composites (MMCs) usually con-sist of a low-density metal, such as aluminum ormagnesium, reinforced with particulate or fibersof a ceramic material, such as silicon carbide orgraphite. Compared with unreinforced metals,MMCs offer higher specific strength and stiffness,higher operating temperature, and greater wearresistance, as well as the opportunity to tailorthese properties for a particular application.

However, MMCs also have some disadvantagescompared with metals. Chief among these are thehigher cost of fabrication for high-performanceMMCs, and lower ductility and toughness. Pres-ently, MMCs tend to cluster around two extremetypes. One consists of very high performancecomposites reinforced with expensive continu-ous fibers and requiring expensive processingmethods. The other consists of relatively low-costand low-performance composites reinforced withrelatively inexpensive particulate or fibers. Thecost of the first type is too high for any but mili-tary or space applications, whereas the cost/ben-efit advantages of the second type over unrein-forced metal alloys remain in doubt.

Current Applications and MarketOpportunities

Current markets for MMCs are primarily in mil-itary and aerospace applications. ExperimentalMMC components have been developed for usein aircraft, satellites, jet engines, missiles, and theNational Aeronautics and Space Administration(NASA) space shuttle. The first production appli-cation of a particulate-reinforced MMC in theUnited States is a set of covers for a missile guid-ance system.

The most important commercial application todate is the MMC diesel engine piston made byToyota. This composite piston offers better wearresistance and high-temperature strength than thecast iron piston it replaced. It is estimated that

300,000 such pistons are produced and sold inJapan annually. This development is very impor-tant because it demonstrates that MMCs are atleast not prohibitively expensive for a very costsensitive application. Other commercial applica-tions include cutting tools and circuit-breakercontacts.

Longer Term Applications

Metal matrix composites with high specific stiff-ness and strength could be used in applicationsin which saving weight is an important factor. In-cluded in this category are robots, high-speed ma-chinery, and high-speed rotating shafts for shipsor land vehicles. Good wear resistance, alongwith high specific strength, also favors MMC usein automotive engine and brake parts. Tailorablecoefficient of thermal expansion and thermal con-ductivity make them good candidates for lasers,precision machinery, and electronic packaging.However, the current level of development ef-fort appears to be inadequate to bring about com-mercialization of any of these in the next 5 years,with the possible exception of diesel enginepistons.

Based on information now in the public do-main, the following military applications forMMCs appear attractive: high-temperature fighteraircraft engines and structures; high-temperaturemissile structures; and spacecraft structures. Test-ing of a National Aerospace Plane (NASP) pro-totype is scheduled for the early to mid 1990s,which might be too early to include MMCs. How-ever, it may be possible to incorporate MMCs inthe structure or engines of the production vehicle.

Research and Development Priorities

MMCs are just beginning to be used in produc-tion applications. In order to make present ma-terials more commercially attractive, and to de-

99

100 . Advanced Materials by Design

velop better materials, the following research anddevelopment priorities should receive attention:

● Cheaper Processes: To develop low-cost,●

highly reliable manufacturing processes, re-search should concentrate on optimizingand evaluating processes such as plasmaspraying, powder metallurgy processes,modified casting techniques, liquid metal in-filtration and diffusion bonding.

● Cheaper Materials: Development of Iower

cost fiber reinforcements is a major need.Continued development work on existingmaterials is important to lower costs as well.Coatings: Research in the area of reinforce-ment/matrix interface coatings is necessary.These coatings can prevent deleterious chem-ical reactions between matrix and reinforce-ment which weaken the composite, particu-larly at high temperature, and optimize theinterracial fiber/matrix bond.

INTRODUCTION

Metal matrix composites (MMCs) generally in dramatic improvements in MMC properties,consist of lightweight metal alloys of aluminum, but costs remain high. Continuously and discon-magnesium, or titanium, reinforced with ceramic tinuously reinforced MMCs have very differentparticulate, whiskers, or fibers.1 The reinforce- applications, and will be treated separatelyment is very important because it determines the throughout this chapter.mechanical properties, cost, and performance of Tailorability is a key advantage of all types ofa given composite. composites, but is particularly so in the case of

Composites reinforced with particulate (dis- MMCs. MMCs can be designed to fulfill require-continuous types of reinforcement) can have ments that no other materials, including other ad-costs comparable to unreinforced metals, with vanced materials, can achieve. There are a num-significantly better hardness, and somewhat bet- ber of niche applications in aerospace structurester stiffness and strength. Continuous reinforce- and electronics that capitalize on this advantage.ment (long fiber or wire reinforcement) can result

PROPERTIES OF METAL MATRIX COMPOSITES

There are considerable differences in publishedproperty data for MMCs. This is partly due to thefact that there are no industry standards forMMCs, as there are for metals. Reinforcementsand composites are typically made by proprie-tary processes, and, as a consequence, the prop-erties of materials having the same nominal com-position can be radically different. The issue isfurther clouded by the fact that many reinforce-ments and MMCs are still in the developmentalstage, and are continually being refined. Numer-ous test methods are used throughout the indus-try, and it is widely recognized that this is a ma-jor source of differences in reported properties.2

1 As used i n this chapter, the terms 1‘al u m inu m, ‘‘magnesium, ”and “titanium” denote alloys of these materials used as matrixmetals.

~Carl Zweben, “Metal Matrix Composites, ” contractor report forOTA, January 1987.

Property data given in this chapter are thereforegiven as ranges rather than as single values.

Some MMC properties cannot be measured asthey would be for monolithic metals. For in-stance, toughness is an important but hard-to-define material property. Standard fracture me-chanics tests and analytical methods for metalsare based on the assumption of self-similar crackextension; i.e., a crack will simply lengthen with-out changing shape. Composites, however, arenon homogeneous materials with complex inter-nal damage patterns. As a result, the applicabil-ity of conventional fracture mechanics to MMCsis controversial, especially for fiber-reinforced ma-terials.

—

Ch. 4—Metal Matrix Composites ● 101

Photo credit: DWA Composite Specialties, Inc.

Micrograph (75x magnification) of 20 percent volumefraction silicon carbide particulate-reinforced

2124 aluminum.

Discontinuous Reinforcement

There are two types of discontinuous reinforce-ment for MMCs: particulate and whiskers. Themost common types of particulate are alumina,boron carbide, silicon carbide, titanium carbide,and tungsten carbide. The most common typeof whisker is silicon carbide, but whiskers of alu-mina and silicon nitride have also been produced.Whiskers generally cost more than particulate,as seen in table 4-1. For instance, silicon carbidewhiskers cost $95 per pound, whereas silicon car-bide particulate costs $3 per pound. Cost pro-jections show that although this difference willdecrease as production volumes increase, par-ticulate will always have a cost advantage.3

Table 4-1 .—Costs of a Representative Sample ofMMC Reinforcements

Reinforcement Price ($/pound)

Alumina-silica fiber. . . . . . . . . . . . . . . . . . . . . 1Silicon carbide particulate . . . . . . . . . . . . . . 3 a

Silicon carbide whisker . . . . . . . . . . . . . . . . . 95Alumina fiber (FP) . . . . . . . . . . . . . . . . . . . . . . 200Boron fiber. . . . . . . . . . . . . . . . . . . . . . . . . . . . $262Graphite fiber (P-100) . . . . . . . . . . . . . . . . . . . 950a

Higher performance reinforcements (e.g., graphite and boron fibers) have sig-nificantly higher costs as well.aJoseph Dolowy, personal communication, DWA Composite Specialties, Inc., JuIy

1987.

SOURCE: Carl Zweben, “Metal Matrix Composites, ” contractor report for OTA,January 1987.

In terms of tailorability, a very important advan-tage in MMC applications, particulate reinforce-ment offers various desirable properties. Boroncarbide and silicon carbide, for instance, arewidely used, inexpensive, commercial abrasivesthat can offer good wear resistance as well as highspecific stiffness. Titanium carbide offers a highmelting point and chemical inertness which aredesirable properties for processing and stabilityin use. Tungsten carbide has high strength andhardness at high temperature.

In composites, a general rule is that mechani-cal properties such as strength and stiffness tendto increase as reinforcement length increases.4

Particulate can be considered to be the limit ofshort fibers. Particulate-reinforced composites areisotropic, having the same mechanical proper-ties in all directions.

In principle, whiskers should confer superiorproperties because of their higher aspect ratio(length divided by diameter). However, whiskersare ‘brittle and tend to break up into shorterlengths during processing. This reduces their rein-forcement efficiency, and makes the much highercost of whisker reinforcement hard to justify. De-velopment of improved processing techniquescould produce whisker-reinforced- MMCs withmechanical properties superior to those madefrom particuiates.

Another disadvantage of using whisker rein-forcement is that whiskers tend to become ori-ented by some processes, such as rolling and ex-trusion, producing composites with differentproperties in different directions (anisotropy).5

41 bid.5See the discussion on an isotropy in ch. 3 on Polymer Matrix Com-

posites.3Ibid

.- —

102 ● Advanced Materials by Design

Anisotropy can be a desirable property, but it isa disadvantage if it cannot be controlled preciselyin the manufacture of the material. It is also moredifficult to pack whiskers than particulate, andthus it is possible to obtain higher reinforce-ment:matrix ratios (fiber volume fraction, v/o)with particulate. Higher reinforcement percent-ages lead to better mechanical properties suchas higher strength.

Continuous Reinforcement

In fiber reinforcement, by far the most com-mon kind of continuous reinforcement, manytypes of fibers are used; most of them are car-bon or ceramic. Carbon types are referred to asgraphite and are based on pitch or polyacryloni-trile (PAN) precursor. Ceramic types include alu-mina, silica, boron, alumina-silica, alumina-boria-silica, zirconia, magnesia, mullite, boron nitride,titanium diboride, silicon carbide, and boron car-bide. All of these fibers are brittle, flaw-sensitivematerials. As such, they exhibit the phenomenonof size effect (see ch. 2); i.e., the strength of thesefibers decreases as the length increases.

Fiber/matrix interface coatings offer anotherdimension of tailorability to MMCs. Coatings arevery important to the behavior of MMCs to pre-vent undesirable reactions, improve the strengthof the fibers, and tailor the bond strength betweenfiber and matrix. A reaction barrier is needed forsome fiber/matrix combinations, particularly

.

Photo credit: DWA Composite Specialties, Inc.

Cross-section of continuous graphite fiber-reinforcedaluminum composite: 52 percent volume fraction fiber,

uniaxially reinforced. 300x magnification

when the composite is exposed to high temper-atures in processing or service. For example, bo-ron fiber can be coated with boron carbide andsilicon carbide reaction barriers to prevent diffu-sion and chemical reactions with the matrix thatdecrease the strength of the composite. Aluminafibers can be given a surface coating of silica toimprove tensile strength.

Coatings can also be used to tailor the bondstrength between fiber and matrix. If adhesionbetween fiber and matrix is too good, cracks inthe matrix propagate right through the fibers, andthe composite is brittle. By reducing the bondstrength, coatings can enhance crack deflectionat the interface, and lead to higher energy ab-sorption during fracture through fiber pulloutmechanisms. Sometimes a coating is needed topromote wetting between the matrix and the fi-ber, and thereby achieve a good bond. Graph-ite can be coated with titanium diboride in or-der to promote wetting.

processing techniques, as well as coatings, canbe used to control deleterious fiber/matrix inter-actions. The application of pressure can be usedto force intimate contact between fiber and ma-trix and thus promote wetting; squeeze castingis one process that does this.

A less common type of continuous reinforce-ment is wire reinforcement. Wires are made ofsuch metals as titanium, tungsten, molybdenum,beryllium, and stainless steel. Such wires offersome tailorability for certain niche applications;for example, tungsten wire offers good high-tem-perature creep resistance, which is an advantagein fighter aircraft jet engines and other aerospaceapplications.

MMC Properties Compared toOther Structural Materials

Table 4-2 compares the most important mate-rial properties of MMCs with those of other struc-tural materials discussed in this assessment.

Strength and Stiffness

The stiffnesses and strengths of particulate-rein-forced aluminum MMCs are significantly betterthan those of the aluminum matrix. For exam-

.

Ch. 4—Metal Matrix Composites ●

pie, at a volume fraction of 40 percent silicon car-bide particulate reinforcement, the strength isabout 65 percent greater than that of the 6061aluminum matrix, and the stiffness is doubled.6

Particulate-reinforced MMCs, which are isotropicmaterials, have lower strength than the axialstrength (parallel to the direction of continuousfiber reinforcement) of advanced polymer matrixcomposites (PMCs), see table 4-2. However, theyhave much better strength than the transversestrength (perpendicular to the direction of con-tinuous fiber reinforcement) of PMCs. The stiff-ness of particulate MMCs can be considered tobe about the same as that of PMCs.

Unlike particulate-reinforced MMCs and mono-lithic metals in general, fiber-reinforced MMCscan be highly an isotropic, having different strengthsand stiffnesses in different directions. The high-est values of strength and stiffness are achievedalong the direction of fiber reinforcement. In thisdirection, strength and stiffness are much higherthan in the unreinforced metal, as shown in ta-ble 4-2. In fact, the stiffness in the axial directioncan be as high as six times that of the matrix ma-terial in a graphite fiber/aluminum matrix com-posite: see table 4-3. However, in the transversedirections, strength values show no improvementover the matrix metal. Transverse strengths andstiffnesses of continuous fiber-reinforced MMCscompared to PMCs are very good, thereby giv-ing MMCs an important advantage over the lead-

bCarl Zweben, “Metal ,Matrlx Composites, ” contractor report fotOTA, January 1987.

103

ing PMCs in structures subject to high transversestresses.

High values of specific strength and specificstiffness (strength and stiffness divided by den-sity) are desirable for high-strength, low-weightapplications such as aircraft structures. Typically,particulate MMCs have somewhat better specificstrength and specific stiffness than the matrixmetal, and fiber-reinforced MMCs have muchbetter specific strength and specific stiffness thanthe matrix metal.

Unfortunately, MMCs have a higher densitythan PMCs, making specific strength and specificstiffness lower than those for PMCs in the axialdirection. Transverse specific strength and spe-cific stiffness of MMCs are still better than thoseof PMCs.

High-Temperature Properties

MMCs offer improved elevated-temperaturestrength and modulus over both PMCs and me-tals. Reinforcements make it possible to extendthe useful temperature range of low density me-tals such as aluminum, which have limited high-temperature capability (see table 4-2). MMCs typi-cally have higher strength and stiffness than PMCsat 200 to 300° C (342 to 5720 F), although de-velopment of resins with higher temperature ca-pabilities may be eroding this advantage. Noother structural material, however, can competewith ceramics at very high temperature.

Fiber-reinforced MMCs experience matrix/rein-forcement interface reactions at high tempera-

Table 4.2.—Structural Properties of Representative MMCs, Compared to Other Materials

Matrix(l)* Particulate(2)* Fiber(3)*Property metala MMC a MMC a PMCb Ceramic b

Strength (M Pa) (axial) . . . . . . . . . . . . . . . . . . . . . . . . . -290Stiffness (G Pa) (axial). . . . . . . . . . . . . . . . . . . . . . . . . - 7 0Specific strength (axial). . . . . . . . . . . . . . . . . . . . . . . -100Specific gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5-2.8Transverse strength (MPa). . . . . . . . . . . . . . . . . . . . . -290Transverse stiffness (GPa) . . . . . . . . . . . . . . . . . . . .Same as axialMaximum use temperature (“C) . . . . . . . . . . . . . . . . 180Plane strain fracture toughness (MPa-m½) . . . . . . 18-35

290-480 620-1,24080-140 130-450

100-170 250-390-2 .8 2.5-3.2

290-480 30-170Same as axial 34-173

300 30012-35 —

820-1,680 140-3,90061-224 97-400

630-670 51-6701.3-2.5 2.7-5.811-56 140-3,9003-12 Same as axial260 1,200-1,600— 3-9

PMC IS used to denote a range of materials including graphite/epoxy, graphite/polylmlde, boron/polyimide, and S-glass/epoxyCeramic is used to denote a range of materials including zirconia, silicon carbide, and silicon nitride“NOTE: (1) 6061 aluminum

(2) 6061 aluminum reinforced with 0-400/0 volume fractions of SiC particulate(3) 6061 aluminum reinforced with 50°/0 volume fractions of fibers of graphite, boron. silicon carbide, or alumina

SOURCES a Carl Zweben, “Metal Matrix Composites, ” contractor report for OTA, January 1987b "Guide to Selecting Engineered Materials. ” Advanced Materials and Processes, vol. 2, No 1, June 1987


Table 4-3.—Strength and Stiffness of SomeFiber. Reinforced MMCs (Fiber V/O = 50%)

Tensile Tensilestrength strength Stiffness Stiffness

Material (axial) (transverse) (axial) (transverse)Matrix material MPa MPa GPa GPa

Aluminum 6061-T6 ...., 290Titanium Ti-6AI-4V 1170

CompositeGraphite a/aluminum . . . 690Boron/aluminum. . . . . 1240Silicon carbideb/

aluminum . . . . . 1240Silicon carbidec/

aluminum . . . . . . . . 1040Alumina d/aluminum . . . . . 620Silicon carbide8/

titanium . . . . . . . . . . . . . 1720Graphite a/copper . . . . . . . 512

2901170

30140

100

70170

34031

70114

450205

205

130205

260464

70114

34140

140

99140

17349

Properly Improvements of MMCs over unreinforced metals can be significant For example, theaxial stiffness of graphite fiber-reinforced aluminum is roughly 6 times greater than that of theunreinforced aluminum The axial tensile strength of silicon carbide fiber-reinforced aluminum IS

about 4 times greater than that of the unreinforced aluminumaP-120bAVCO monofilamentc"Nicalon"dFibe r "FP"

‘AVCO monofilament

SOURCES: For all composites except graophit/copper: Carl Zweben, “Metal Matrix Composites, ”contractor report for OTA, January 1987. For graphite/copper only: C Kaufmann, Ameri-can Cyanamid, personal communication, Oct. 19, 1987.

tures. In addition, the transverse high-tempera-ture strength of fiber-reinforced MMCs is only asgood as that of the matrix metal, since mechani-cal properties in the transverse direction are dom-inated by the matrix and the fiber/matrix inter-face. For example, at 320 C (608° F), the axialtensile strength of boron fiber-reinforced alumi-num is about 1.1 gigapascals (GPa) compared toonly 0.07 GPa for monolithic 6061 aluminum,whereas the transverse strength is 0.08 GPa,about the same as that for the monolithic 6061.7

Wear Resistance

Wear resistance of MMCs is excellent comparedto that of monolithic metals and PMCs, owing tothe presence of the hard ceramic reinforcements.For instance, in one test, the abrasive wear of2024 aluminum under a 1 kilogram load wasshown to be 6 times greater than the wear of thesame alloy containing 20 percent volume frac-tion of silicon carbide whiskers.8 An alumina-silicafiber-reinforced aluminum piston used in Toyota

automobiles demonstrated an 85 percent improve-ment in wear resistance over the cast iron pis-ton with nickel insert used previously.9

Fracture and Toughness

There is a wide variation in fracture toughnessamong MMCs, although it is generally lower thanthat of the monolithic metal. Fracture toughnesscan vary between 65 and 100 percent of the frac-ture toughness of the monolithic metal alloy.10

Lower toughness is a trade-off for higher strengthand stiffness. Particulate-reinforced MMCs havea lower ultimate tensile strain than the unrein-forced metals (see table 4-4) which may be im-portant in some applications. This brittleness cancomplicate the design process and make joiningmore difficult as well. Comparison to PMCs is dif-ficult, because the toughness of PMCs is verytemperature-dependent.

Thermal Properties

The introduction of silicon carbide particulateinto aluminum results in materials having lowercoefficients of thermal expansion, a desirableproperty for some types of applications. By choos-ing an appropriate composition, the coefficient

7 Ibid.

8 lbid.9 lbid.10 lbid.

Ch. 4—Metal Matrix Composites . 105

Table 4-4.—Properties of 6061 Aluminum ReinforcedWith Silicon Carbide Particulate

TensileStiffness strength Tensile

Reinforcement (G Pa) (M Pa) strain

0% 10 290 16.O%10 83 7.015 86 3&l 6.520 90 430 6.025 103 — 3.530 117 470 2.035 124 — 1.040 140 480 0.6

Note the sharp rise in stiffness in the 400/0 composite (140 GPa) compared tothe unreinforced aluminum (10 GPa). Ultimate tensile strength nearly doublesand tensile strain decreases nearly a full order of magnitude.

SOURCE: Carl Zweben, “Metal Matrix Composites,” contractor report for OTA,January 1987.

of thermal expansion can be near zero in someMMCs. MMCs also tend to be good heat conduc-tors. Using high thermal conductivity graphitefibers, aluminum-matrix or copper-matrix MMCscan have very high thermal conductivity, com-pared with other types of composites.

Environmental Behavior

In terms of environmental stability, MMCs havetwo advantages over PMCs. First, they suffer less

water damage than PMCs which can absorbmoisture, thereby reducing their high-tempera-ture performance. Second, some MMCs, such asreinforced titanium, can stand high-temperaturecorrosive environments, unlike PMCs.

Nevertheless, some MMCs are subject to envi-ronmental degradation not found in PMCs. Forinstance, graphite fibers undergo a galvanic re-action with aluminum. This can be a problemwhen the graphite/aluminum interface is exposedto air or moisture. In addition, PMCs are resis-tant to attack by many chemicals (e.g., acids) thatcorrode aluminum, steel, and magnesium.

cos t

A major disadvantage of MMCs compared tomost other structural materials is that they aregenerally more expensive. Both constituent ma-terial costs and processing costs are higher. Costsof higher performance reinforcements (mostlyfibers) are high, and lower cost reinforcements(mostly particulate) may not yield dramatic im-provements in performance. Cost/benefit ratiosof most MMCs dictate that they be used only inhigh-performance applications. However, both

MICROWAVE CIRCUIT PACKAGING

KOVAR

WEIGHT = 42

THERMALCONDUCTIVITY

METAL MATRIX COMPOSITE

g WEIGHT = 15 g

9.6 BTU THERMAL 74 BTU=HR-FT-°F CONDUCTIVITY = HR-FT-°F

Photo credit: General Electric Co.

Silicon carbide particle-reinforced aluminum microwave package


the material and production costs may drop asmore experience is gained in MMC production.

Tailorability of Properties

In some applications, MMCs offer unique com-binations of properties that cannot be found inother materials. Electronics packaging for aircraftrequires a hard-to-achieve combination of lowcoefficient of thermal expansion, high thermalconductivity, and low density. Certain MMCs canmeet these requirements, replacing beryllium,which is scarce and presents toxicity problems.

Several other examples can be cited to illus-trate the unique advantages of MMCs in specialtyapplications. A large heat transfer coefficient isdesirable for space-based radiators; this propertyis offered by graphite fiber-reinforced copper, alu-minum, and titanium (although this latter fiber/matrix combination unfortunately has some in-terface reaction problems). The higher transversestrengths of MMCs compared to PMCs have ledto several MMC space applications, such as thespace shuttle orbiter struts, made of boron fiber-reinforced aluminum.

One of the most important applications ofMMCs is the Toyota truck diesel engine piston,produced at a rate of about 300,000 pistons perYear.11 This consists of aluminum selectively rein-forced in the critical region of the top ring groovewith a ring-shaped ceramic fiber preform. (A pre-form is an assemblage of reinforcements in theshape of the final product that can be infiltratedwith the matrix to form a composite.)

Two types of fibers are used: alumina andalumina-silica. Both are relatively low-cost ma-terials originally developed for furnace insulation.

11 Ibid. —

Photo credit: Toyota Motor Corp.

Aluminum diesel engine piston with local fiberreinforcement in ring groove area

Use of local reinforcement to improve wear re-sistance makes possible the elimination of nickeland cast iron inserts. This reduces piston weightand increases thermal conductivity, improvingengine performance and reducing vibration.

This approach minimizes cost and thereby makesMMCs more competitive with cast iron. This de-sign not only offers better wear resistance andbetter high-temperature strength but also elimi-nates one type of part failure associated with thedesign it replaced. It is considered by some tobe a development of historic proportions becauseit demonstrates that MMCs can be reliably massproduced. It also shows that at least one type ofaluminum matrix MMC can perform reliably ina very severe environment.

DESIGN, PROCESSING, AND TESTING

In ceramics and polymer matrix composites billets or sheets and later machined to a finalmanufacturing, the material is formed to its final shape. Viable, inexpensive near-net-shape tech-shape as the microstructure of the material is niques for forming MMCs have not been success-formed. MMCs can be produced in this way, or, fully developed as yet, and there is still a debateas is traditionally done with metals, formed into about on the advantages of producing standard


Photo credit: AVCO Corp.

Centrifugal casting of silicon carbide-reinforcedaluminum.

shapes to be machined by the purchaser to finalform, compared to the custom production ofnear-net-shape parts. Forming of net-shape MMCparts necessitates close integration of design andmanufacture, where production of standard MMCstructural shapes does not.

Design

The variability of metal matrix composite prop-erties is a handicap to designing with these ma-terials. There are currently no design aids suchas property databases, performance standards,and standard test methods. In addition, mostdesigners are much more familiar with monolithicmetals. The lack of experience with MMCs in-creases the design and manufacturing costs asso-ciated with the development of a new product.As with ceramic structures, brittleness is a newand difficult concept to most designers. MMCsare brittle materials, and new methods of designwill become important in the development ofMMC applications.

Processing

Because MMC technology is hardly beyond thestage of R&D at present, costs for all methods ofproducing MMCs are still high. Manufacturingmethods must ensure good bonding between ma-trix and reinforcement, and must not result in un-desirable matrix/fiber interracial reactions.

MMC production processes can be divided intoprimary and secondary processing methods,though these categories are not as distinct as isthe case with monolithic metals. Primary proc-esses (those processes used first to form the ma-terial) can be broken down into combining andconsolidation operations. Secondary processescan be either shaping or joining operations. Ta-ble 4-5 shows the manufacturing methods dis-cussed here and notes which types of operationsare included in each method.

As with ceramics, net-shape methods are veryimportant manufacturing processes. The machin-ing of MMCs is very difficult and costly in thatMMCs are very abrasive, and diamond tools areneeded. In addition, it is desirable to reduce theamount of scrap left from the machining proc-ess because the materials themselves are very ex-pensive.

Continuous Reinforcement12

The following discussion covers the primaryand secondary processes involved in the manu-facture of MMCs with continuous reinforcement.

Primary Processes.– Basic methods of combin-ing and consolidating MMCs include liquid metalinfiltration, modified casting processes, and depo-sition methods such as plasma spraying. Hotpressing consolidates and shapes MMCs. Diffu-sion bonding consolidates, shapes and joinsMMCs. See table 4-6 for selected MMC process-ing methods and their characteristics.

There is considerably more disagreement onwhat is the most promising method for process-ing fiber-reinforced MMCs than there is for proc-essing of particulate-reinforced MMCs. The Toyotadiesel engine piston has been heralded as an ex-ample of the promise of modified casting tech-niques for keeping costs down. Critics charge thatthese types of casting techniques have not provenadequate for manufacturing MMCs with desira-ble mechanical properties. (The Toyota piston

12 As examples in the following discussion of processing, two com-

mon types of particulate- and fiber-reinforced composites will bereferred to as needed: silicon carbide particulate-reinforced alu-minum, and boron fiber-reinforced alum inure.

108 ● Advanced Materials by D e s i g n

Table 4-5.—MMC Manufacturing Methods

C o m b i n e s C o n s o l i d a t e s S h a p e s J o i n s

Primary methods:Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x x

Squeeze casting,compocast ing,gravity casting,low-pressure casting

Diffusion bonding . . . . . . . . . . . . . . . . . . . . . . . . . . x x x

Liquid infiltration . . . . . . . . . . . . . . . . . . . . . . . . . . . . x xGravity,inert gas pressure,vacuum infiltration

Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X xChemical coating,plasma spraying,chemical vapor deposition,physical vapor deposition,electrochemical plating

Powder processing . . . . . . . . . . . . . . . . . . . . . . . . . . X x xHot pressing,ball mill mixing,vacuum pressing,extrusion, rolling—

Secondary methods:Shaping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x

Forging, extruding, rolling,bending, shearing, spinning,machining

Machining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xTurning, boring,drilling, milling,sawing, grinding, routing,electrical discharge machiningchemical milling,electrochemical milling

Forming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x x xPress brake,superplastic,creep forming

Bonding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xAdhesive, diffusion

Fastening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xSoldering, brazing, welding . . . . . . . . . . . . . . . . . . . xSOURCE: Carl Zweben, “Metal Matrix Composites, ” contractor report for OTA, January 1987.

uses fiber reinforcement mostly for wear resis-tance and less for more universally importantproperties such as strength.) Some industry MMCadvocates suggest that processes such as diffu-sion bonding and plasma spraying are more prom-ising for achieving high performance, and thatlower costs will come only with more produc-tion experience with these processes.

secondary Processes.–Machining processesfor MMCs reinforced with ceramic materials, which

expensive than for monolithic structural metals.As a rule, diamond tools are required becausecarbide and other tools wear out too quickly. De-spite this limitation, all of the basic mechanicalmethods, such as drilling, sawing, milling, andturning have been proven to be effective withMMCs. Electrical discharge machining also hasbeen shown to be effective.

Mechanical fastening methods, such as rivet-ing and bolting have been found to work for

are hard and abrasive, generally are much more continuously-reinforced MMCs. The same is true

Ch. 4—Metal Matr ix Composi tes ● 1 0 9

Table 4-6.—Selected MMC Processing Techniquesand Their Characteristics

Techniques Characteristics

For fiber-reinforced MMCs:Liquid metal infiltration near-net-shape parts

(low pressure) economicalhigh porosityoxidation of matrix and fibernot reliable as yet

Liquid metal infiltration (intert gas less porosity and oxidation than low pressurepressure, vacuum) techniques

Squeeze casting

Low pressure casting near-net-shape partslow costexpensive preforms requiredthree-dimensional preforms are difficult to make

good fiber wettinglower porosityexpensive molds neededlarge capacity presses needed

Plasma spraying

Diffusion bonding

potential for lower processing costs

lower temperatures than hot pressingreduces fiber/matrix interactionsnot capable of net-shape parts except simple

shapesslow, expensivefiber damage can occur

Hot Pressing heats matrix above melt temperature, which candegrade reinforcement

For particulate-reinforced MMCs:Powder metal lurgy high volume fractions of particulate are possible

(better properties)powders are expensivenot for near-net-shape parts

Liquid metal inf i l trat ion net-shape partscan use ingots rather than powderslower volume fraction of particulate (means

lower mechanical properties)

Squeeze casting may offer cost advantage; however molds andpresses may be expensive

SOURCE Carl Zweben, “Metal Matrix Composites, ” contractor report for OTA, January 1987

for adhesive bonding. Boron fiber-reinforced alu-minum pieces can be metallurgically joined bysoldering, brazing, resistance welding and diffu-sion bonding.

Discontinuous Reinforcement

The following discussion covers the primaryand secondary processes involved in the manufac-ture of MMCs with discontinuous reinforcement.

Primary Processes.–The most common meth-ods for producing particulate- and whisker-rein-

forced MMCs are powder processing techniques.There has not been much success at producingnear-net-shape structures as yet. 13 (See discussionin chapter 2 on the desirabiIity of near-net-shapeprocessing.) Other processes for discontinuously-reinforced MMCs are Iiquid metal infiltration andcasting techniques (see table 4-6).

At this time, there is no one MMC manufac-turing method that holds great promise for reduc-


Photo credit: AVCO Specialty Materials

Centrifugal casting around braided tube of SiCfibers and yarn.


Diffusion bonded SiC fiber-reinforced titanium:cross section of a hollow turbine blade.

ing costs, although there seems to be some agree-ment that for particulate-reinforced MMCs, liquidmetal infiltration and powder metallurgy techniquesare likely candidates for future development.

Secondary Processes.–One of the major ad-vantages of particulate- and whisker-reinforcedMMCs is that most of the conventional metal-working processes can be used with minor mod-ifications (see table 4-5), Methods demonstratedinclude forging, extruding, rolling, bending,shearing, and machining. Although they are ex-pensive, all conventional machining methodshave been found to work for these materials, in-cluding turning, milling, and grinding.


Silicon carbide fiber-reinforced aluminum shaped byelectrical discharge machining.

A wide variety of joining methods can be usedincluding mechanical fastening techniques, met-allurgical methods employed with monolithicmetals, and adhesive bonding as used for PMCs.

costs

MMC costs are currently very high, and forthem to come down, there must be some stand-ardization and a reliable compilation of materi-als properties (in a form such as a design data-base). This can be achieved only through greaterproduction experience with MMC materials.Some experts argue that entirely new materialsand processes must be found that are cheaperthan the present ones. Opponents of this viewhold that development of new processes and ma-terials will delay a decrease in costs of existingmaterials, because it will take much longer to gainthe production experience necessary.

Most present users buy preformed shapes, suchas billets, plates, bars and tubes, and then ma-chine these shapes to specification. MMCs arenot currently sold in standard sizes; users can or-der any size, manufactured to order. This lackof standardization keeps prices for shapes high.Some users produce their own MMCs for use in-house. For example, Lockheed Georgia producessilicon carbide fiber-reinforced aluminum matrixcomposites for designing, manufacturing and test-ing fighter aircraft fins.


The cost of an MMC part depends on manyfactors including shape, type of matrix and rein-forcement, reinforcement volume fraction, rein-forcement orientation, primary and secondaryfabrication methods, tooling costs, and numberof parts in the production run. At present, thereis little information available on production costs.The only items produced in any quantity are theToyota engine pistons. Costs for this applicationare proprietary information, but the costs of theseMMCs are at least not prohibitive in a very cost-sensitive product.

Costs are very volume sensitive; high costs keepvolumes low, and low volumes mean high costs.Of course, some materials are inherently expen-sive and will never be cheap enough for wide-spread commercial use, regardless of volume.

With the exception of alumina-silica discontinu-ous reinforcement fibers, reinforcement pricesare orders of magnitude higher than those of me-tals used in mass production items such as au-tomobiles. Unfortunately, the stiffness of alumina-silica fibers is not substantially greater than thatof aluminum. The room-temperature strength ofaluminum reinforced with these fibers also showslittle or no improvement over monolithic alumi-num, although wear resistance and elevated-temperature strength are enhanced. The signifi-cance of this is that the fibers that provide majorimprovements in material properties are quite ex-pensive, while use of low-cost alumina-silicafibers provides only modest gains in strength andstiffness.

Testing

Unlike monolithic metals, in which the maintypes of flaws are cracks, porosity, and inclusions,

composites are complex, heterogeneous mate-rials that are susceptible to more kinds of flaws,including delamination, fiber misalignment, andfiber fracture.

Failure in monolithic metals occurs primarilyby crack propagation. The analytical tool calledlinear elastic fracture mechanics (LEFM) is usedto predict the stress levels at which cracks inmetals will propagate unstably, causing failure.Failure modes in composites, especially thosereinforced with fibers or whiskers, are far morecomplex, and there are no verified analyticalmethods to predict failure stress levels associatedwith observed defects. As a consequence, em-pirical methods have to be used to evaluate howcritical a given flaw is in an MMC.

Two basic methods have been established asreliable flaw-detection techniques for MMCs:radiography and ultrasonic C-scan. Radiography,useful only for thin panels, detects fiber misalign-ment and fractures. C-scan identifies delamina-tion and voids. There are no existing nondestruc-tive evaluation (N DE) methods for reinforcementdegradation in MMCs.

The costs of NDE methods for MMCs shouldbe no greater than for monolithic metals. How-ever, the reliability of manufacturing processesfor these materials has not been established andMMCs cannot be reliably or repeatably producedas yet. Because fabrication processes are not de-pendable at present, it is generally necessary touse NDE for MMCs in cases where testing wouldnot be employed at all, or as extensively, formonolithic metals, This additional cost factorshould decrease as experience and confidenceare gained with MMCs.

HEALTH AND SAFETY

There is little documentation as yet of safe han- Health hazards associated with MMCs are sim-dling and machining practices for MMCs. Mate- ilar to those found in the production of ceram-rials handling practices are given by materials ics and PMCs. As with ceramics the most serioussafety data sheets, and few yet exist for MMCs.However, there is one materials safety data sheetthat applies to the MMC production process of lqRObefl BUffenbarger, I rlternational Assoclatlon of Machinists andplasma spray ing.14 Aerospace Workers, personal communication, March 23, 1987.


health hazard is associated with the possible car- boria-silica. The effects of ceramic fibers on hu-cinogenic effects of ceramic fibers due to their mans are largely unknown. Until such data be-size and shape. MMC reinforcement fibers fall- come available, the indications discussed in chap-ing in this category include alumina, alumina- ter 2 for ceramics also apply to the manufacturesilica, graphite, boron carbide, and alumina- of MMCs reinforced with these fibers.

APPLICATIONS

There are very few commercial applications ofMMCs at the present time. Industry observers be-lieve that the level of development effort in theUnited States is not large enough to lead to sig-nificant near-term commercial use of MMCs.However, the Toyota truck diesel engine pistonhas spurred the major U.S. automakers to under-take preliminary activity in MMCs. Although con-siderable interest in these materials is being gen-erated in the United States, it will be decadesbefore they are likely to have an appreciable im-pact on production levels of competing materials.

MMCs are not competing solely with mono-lithic metals; they are also competing with thewhole range of advanced materials, includingPMCs, ceramics, and other new metal alloys. Itis not yet clear whether, compared to PMCs, ce-ramics, and monolithic metals, MMCs will havebig enough performance advantages to warrantuse in a wide variety of applications. In fact,MMCs may be limited to niche applications inwhich the combinations of properties requiredcannot be satisfied by other structural materials.

The properties that make MMCs attractive arehigh strength and stiffness, good wear resistance,and tailorable coefficient of thermal expansionand thermal conductivity. Furthermore, develop-ment of high-temperature MMCs has been citedas a way to help reduce U.S. dependence on crit-ical and strategic materials, such as manganeseand cobalt. ’ 5 The following sections describe thecurrent and future applications for which theseproperties are of value.

Current Applications

Current MMC markets in the United States areprimarily military. MMCs have potential for usein aircraft structures, aircraft engines, and navalweapons systems. Experimental MMC compo-nents were first developed for applications in air-craft, jet engines, rockets, and the space shuttle.There has been little government funding ofMMCs for commercial applications in the UnitedStates, and only a small number of specializedapplications have been developed by private in-dustry.

Military/Space

The high stiffness and compression strength andlow density of boron fiber-reinforced aluminumled to its use in production space shuttle orbiterstruts. Its high cost has prevented wider use.

The National Aeronautics and Space Adminis-tration (NASA) Hubble Space Telescope uses twoantenna masts made of aluminum reinforced withP-100 pitch-based graphite fibers. This materialwas selected because of its high specific stiffnessand low coefficient of thermal expansion. 16 Thismaterial replaced a dimensionally stable telescopemount and an aluminum waveguide, resulting ina 70 percent weight savings. 17

Two companies are producing silicon carbideparticulate-reinforced aluminum instrument cov-ers for a missile guidance system. This is the firstproduction application of particulate-reinforcedMMCsin the United States.

IJzweben, op. cit., January 1987

161bid.

I Twilllam C, Riley, Research Opportu nites, Inc., personal com-

munication, October 27, 1987.


AIRFOIL CROSS SECTION

X75 X100

Photo credit: NASA Lewis Research Center

Tungsten wire-reinforced super alloy gas turbineengine blade

Commercial

The Toyota piston has stimulated a consider-able interest in MMCs for pistons and other au-tomotive parts on a worldwide basis. Other com-ponents now being evaluated by automakersworld-wide include connecting rods, cam fol-lowers, cylinder liners, brake parts, and driveshafts.

Experimental MMC bearings have been testedon railroad cars in the United Kingdom. An ex-perimental material for electronic applications hasbeen developed in Japan, consisting of copperreinforced with graphite fiber, Another electricaluse of MMCs is in circuit breakers. Graphite-reinforced copper used as a “compliant layer”minimizes thermal stresses in an experimentalmagnetohydrodynamic (MHD) generator ceramicchannel in Japan. An aluminum tennis racketselectively reinforced with boron fiber was soldcommercially in the United States for a brief timeduring the 1970s,

Future Applications

Future applications include those which couldbe possible in the next 5 to 15 years.

Photo credit: Advanced Composite Materials

Missile guidance system covers of silicon carbideparticle-reinforced aluminum.

Military/Space

Based on information in the public domain, thefollowing applications appear likely to be devel-oped in the United States: high-temperaturefighter aircraft engines and structures, high-tem-perature missile structures, spacecraft structures,high-speed mechanical systems, and electronicpackaging.

Development of applications such as hyper-sonic aircraft, which will require efficient high-temperature structural materials, will undoubt-edly lead to increasing interest in MMCs.

Commercial

Commercial applications in the next 5 years arelikely to be limited to diesel engine pistons, andperhaps sporting goods such as golf clubs, tennisrackets, skis, and fishing poles. There are a num-

.—— .


ber of potential applications that are technicallypossible but for which the current level of devel-opment effort is inadequate to bring about com-mercialization in the next 5 years; these include:brake components, push rods, rocker arms, ma-chinery and robot components, computer equip-ment, prosthetics, and electronic packaging.

High Specific Stiffness and Strength.–MMCmaterials with high specific stiffness and strengthare likely to have special merit in applications inwhich weight will be a critical factor. Includedin this category are land-based vehicles, aircraft,ships, machinery in which parts experience highaccelerations and decelerations, high-speed shaftsand rotating devices subject to strong centrifu-gal forces.

The high value of weight in aircraft makes useof MMCs a strong possibility for structural appli-cations, as well as for engine and other mechan-ical system components. Mechanical propertiesof MMCs could be important for a number ofmedical applications, including replacement joints,bone splices, prosthetics, and wheelchairs.

There are numerous industrial machinery com-ponents that could benefit from the superior spe-cific strength and stiffness of MMCs. For exam-ple, high-speed packaging machines typicallyhave reciprocating parts. Some types of high-speed machine tools have been developed to thepoint where the limiting factor is the mass of theassemblies holding the cutting or grinding tools,which experience rapid accelerations and de-celerations. Further productivity improvementswill require materials with higher specific me-chanical properties. Computer peripheral equip-ment, such as printers, tape drives, and magneticdisk devices commonly have components thatmust move rapidly. MMCs are well suited forsuch parts.

The excellent mechanical properties of MMCsmake them prime candidates for future applica-tion in robots, in which the weight and inertiaof the components have a major effect on per-formance and load capacity. Centrifugal forcesare a major design consideration in high-speedrotating equipment, such as centrifuges, gener-ators, and turbines. As these forces are directlyproportional to the mass of the rotating compo-

nents, use of materials with high specific prop-erties - MMCs - will be a way of achieving futureperformance goals.

For high-speed rotating shafts, such as automo-bile and truck drive shafts, a major design con-sideration is that the rotational speed at whichthe shaft starts to vibrate unstably, called the crit-ical speed, must be higher than the operatingspeed. As critical speed depends on the ratio ofstiffness to density, the high specific stiffness ofMMCs makes them attractive candidates for thisapplication.

Attractive Thermal Properties. -There are anumber of potential future applications for whichthe unique combinations of physical propertiesof MMCs will be advantageous. For example, thespecial needs of electronic components presentparticularly attractive opportunities.

Electronic devices use many ceramic and ce-ramic-like materials, that are brittle and have lowcoefficients of thermal expansion (CTEs). Exam-ples are ceramic substrates such as alumina andberyllia, and semiconductors such as silicon andgallium arsenide. These components frequentlyare housed in small, metallic packages. MMCscan help prevent fracture of the components orfailure of the solder or adhesive used to mountthese components, since the CTE of the packagecan be tailored to match that of the device.


Another des i rab le feature of packaging mate-

rials is high thermal conductivity to dissipate heat

generated in the system, since the reliability ofelectronic chips decreases as operating temper-ature increases. Two of the more common me-tals now used in packaging are molybdenum andKovar, a nickel-iron alloy. The thermal conduc-tivity of Kovar is quite low, only about 5 percentof that of copper, and it cannot be used in appli-cations in which large amounts of heat must bedissipated. Molybdenum, which is more expen-sive and difficuIt to machine, is normally used insuch cases. MMCs (e. g., with a high thermal con-ductivity copper matrix) could be used insteadof these two materials. Examples of potential fu-ture applications include heat sinks, power semi-conductor electrodes, and microwave carriers.

The low CTE that can be achieved with someMMCs makes them attractive for use in precisionmachinery that undergoes significant temperaturechange. For example, machines that assembleprecision devices are commonly aligned whencold. After the machines warm up, thermal ex-pansion frequently causes them to go out of align-ment. This results in defective parts and downtime required for realignment. Laser devices,which require extremely stable cavity lengths, areanother potential application for which low CTEis an advantage.

Markets

Market projections for MMCs vary widely. Onemarket forecast by C. H. Kline is that MMCs willplay no significant role in the current advancedcomposite markets until after 1995.18 A secondforecast (Technomic Consultants) is that U.S. non-military uses of MMCs will reach $100 million peryear by 1994, and world-wide commercial useswill reach $2 billion per year. 19

There does seem to be agreement though, thatin the United States, MMC materials are likely tobe used primarily in military and space applica-tions, and, to a much smaller degree, in electronicand automotive applications. Because MMC ma-terials are currently used mainly for research pur-poses, the markets for them are now only a fewthousand pounds per year.20 Materials costs shoulddecrease with time, as production volumes in-crease. However, the actual downward trend hasbeen slower than most predictions, and the cost/performance benefits of MMCs have yet to bedemonstrated over alternative materials in largevolume applications.

—.. —lgjacques shou~ens, Metal Matrix Composites Information Anal-

ysis Center (MMCIAC), personal communicat ion, March 27, 1987.lglbid.

ZOI bid.

RESEARCH AND DEVELOPMENT PRIORITIESFederal funding of MMC research and devel-

opment (R&D) comes mainly from the military.Combined totals for the three services plus theDefense Advanced Research Projects Agency(DARPA) for 6.1,6.2, and 6.3A money were $29.7million for fiscal year 1987.21 There are a num-ber of classified projects, and projects in othercategories of funds, that also involve MMCs. Al-though the Department of Defense is the majorgovernment funding source, NASA also fundsMMC research. NASA plans to provide $8.6 mil-lion for MMCs in fiscal year 1988, up from $5.6million in fiscal year 1987.22

~ 1 Jerome” Persh, Department of Defense, personal comm u nlca-

tlon, August 1987.~~Brlan Qulgley, NatlOndl Aeronautics and Space Admin is t ra t ion,

p e r s o n a l c o m m u n i c a t i o n , D e c e m b e r 1 9 8 7 .

The following section describes R&D prioritiesfor MMCs in three broad categories of descend-ing priority. These categories reflect a consensusas determined by OTA of research that needs tobe done in order to promote the developmentof these materials.

Very Important

Cheaper Processes

First in importance is the need for low-cost,highly-reliable manufacturing processes. Severalindustry experts advocate emphasis on modifiedcasting processes; others suggest diffusion bond-ing and liquid metal infiltration as likely candi-dates for production of fiber-reinforced MMCs.


There is some agreement that p lasma spray ing

is also a promising method. For particulate-rein-forced MMCs, powder metallurgy and liquid in-filtration techniques are considered most prom-ising. To develop near-term applications, researchshould concentrate on optimizing and evaluat-ing these processes, including development oflow-cost preforms.

Cheaper Materials

Development of high-strength fiber reinforce-ments of significantly lower cost is a major need,as there are no high-performance, low-cost fibersavailable at present. There are two schools ofthought as to the best approach to reducing costs.Some analysts believe that the range of useful-ness can be expanded by the development ofnew fibers with better all-purpose design prop-erties, such as higher strength, higher tempera-ture, and lower cost; they also see new matrixalloys as important to facilitate processing and tooptimize MMC performance. However, somecritics charge that this search for all-purpose fibersand better matrices is likely to be unrewardingand that increased production experience withpresent fibers and matrices is the best route tolower costs, particularly for potential near-termapplicat ions.

Interphase

Control over the fiber/matrix interphase inMMCs is critical to both the cost and perform-ance of these materials. For example, new fibercoatings are needed to permit a single fiber tobe used with a variety of matrices. This wouldbe cheaper than developing a new fiber for eachmatrix.

At high temperatures, fiber/matrix interactionscan seriously degrade MMC strength. Researchin the area of coatings is desirable, not only toprevent these deleterious reactions but also topromote the proper degree of wetting to form agood fiber/matrix bond. Coatings add to the ma-terial and processing costs of MMCs, so that re-search into cheaper coatings and processes is es-sential.

Important

Environmental Behavior

It is critical to understand how reinforcementsand matrices interact, particularly at high tem-peratures, both during fabrication and in service.A thorough understanding of material behavioris necessary to ensure reliable use and to provideguidelines for development of materials with im-proved properties. To date, study of the behaviorof MMCs in deleterious environments has beenmuch more limited than for PMCs. Research isneeded in the areas of stress/temperature/defor-mation relations, strength, mechanical and ther-mal fatigue, impact, fracture, creep rupture, andwear.

Fracture

The subject of fracture behavior in MMCsdeserves special attention. The applicability oftraditional analytical techniques is controversialbecause MMCs are strongly heterogeneous ma-terials. It seems reasonable that these traditionaltechniques should be valid at least for particulate-reinforced MMCs, because they do not appearto have the complex internal failure modes asso-ciated with fiber-reinforced MMCs. In view of thelack of agreement on how to characterize frac-ture behavior of MMCs, though, it will be nec-essary to rely on empirical methods until a clearerpicture emerges.

Nondestructive Evaluation

Reliable NDE techniques must be developedfor MMCs. They should include techniques fordetecting flaws and analytical methods to evalu-ate the significance of these flaws. In MMCs asin PMCs, there is the possibility of many kindsof flaws, including delamination, fiber misalign-ment, and fracture. There are no N DE procedurespresently available for measuring the extent ofundesirable reinforcement/matrix interaction andresultant property degradation.

Machining

The machining of MMCs is currently a very ex-pensive process that requires diamond tools be-

.-


cause of the materials’ very hard and abrasive ce-ramic reinforcements. As machining is a majorcost factor, development of improved methodstailored to the unique properties of MMCs wouldhelp to make these materials more competitive.

Desirable

Modeling

fabricating MMCs. What are needed are designmethods that take into account plasticity effectsand provide for development of efficient, selec-tively reinforced structures and mechanical com-ponents. Research on modeling of processing be-havior is necessary, together with the eventualdevelopment of databases of properties and proc-ess parameters.

Analytical modeling methods would be of valuein helping to develop and optimize processes for

Documents

Metal Matrix Composites