Volume 18 - Friction, Lubrication, and Wear Technology Part 21 pdf

This arbitrary division is based on the fact that composites containing hard particles generally exhibit different friction and wear behavior because of different wear mechanisms than co

Fig 3 Microstructure of type A390.0 hypereutectic alloy (a) Unrefined (Graff-Sargent etch) Dark regions

contain coarse primary silicon particles in addition to eutectic silica (b) Refined (as polished) 120×

Commercial aluminum-silicon alloys (Table 1) generally contain other alloying elements to further enhance or modify the wear resistance or impart additional properties to these alloys

Iron. The most common alloying element is iron, which can be tolerated up to levels of 1.5 to 2.0% Fe The presence of

iron modifies the silicon phase by introducing several Al-Fe-Si phases The most common of these are the and phases The phase has a cubic crystal structure and appears in the microstructure as a "Chinese script" eutectic The less common phases generally appear as needles and/or platelets in the structure Other iron-bearing phases such as Al6Fe and FeAl3 can also be found in these alloys Aluminum-silicon alloys intended for die castings typically have higher minimum iron levels to reduce sticking between the mold and the casting

Magnesium is added to provide strengthening through precipitation of Mg2Si in the matrix In an Al-Fe-Si-Mg alloy, the Al-Si-Fe phases will not be affected by the addition of magnesium However, magnesium can combine with insoluble aluminum-iron phases, resulting in a loss of strengthening potential (Ref 12)

Copper. The most common aluminum wear-resistant alloys also contain copper Copper additions impart additional strengthening of the matrix through the aging or precipitation-hardening process (AlCu2 or Q phase) or through modification of the hard, brittle Al-Fe-Si phases by substitution in these intermetallic phases As the strength of these alloys increases through magnesium and copper additions, some sacrifice in ductility and corrosion resistance occurs

Manganese. Many of the important aluminum-silicon alloys also contain low (<1 wt%), but significant, amounts of manganese The presence of manganese can reduce the solubility of iron and silicon in aluminum and alter the composition and morphology of the Al-Fe-Si primary constituent phases For example, manganese additions can favor the formation of constituents such as Al12 (Fe,Mn)3 rather than the Al9Fe2Si2-type constituents The manganese-bearing constituents are typically less needlelike or platelike than the manganese-free iron- or (iron/silicon)-bearing primary constituents Manganese additions also improve elevated-temperature properties of the aluminum-silicon alloys

Cumulative Effect of Alloying Elements. In summary, aluminum wear-resistant alloys are based on alloys containing the land, brittle silicon phase Alloying elements such as iron, manganese, and copper increase the volume fraction of the intermetallic silicon-bearing phases, contributing to increased wear resistance compared to binary aluminum-silicon alloys In addition, magnesium and copper also provide additional strengthening by producing submicroscopic precipitates within the matrix through an age-hardening process

Properties and Structure

Alloying aluminum with silicon at levels between about 5 and 20% imparts a significant improvement in the casting characteristics relative to other aluminum alloys As a result, these high-silicon alloys are generally utilized as casting alloys rather than for the manufacture of wrought products Aluminum-silicon alloys also possess excellent corrosion resistance, machinability, and weldability (Table 3)

Table 3 Relative ratings of aluminum-silicon sand casting and permanent mold casting alloys in terms of castability, corrosion-resistance, machinability, and weldability properties

Fluidity (c) Shrinkage

tendency (d)

Resistance

to corrosion (e)

(a) For ratings of characteristics, 1 is the best and 5 is the poorest of the alloys listed Individual alloys may have

different ratings for other casting processes

(b) Ability of alloy to withstand stresses from contraction while cooling through hot-short or brittle temperature

range

(c) Ability of molten alloy to flow readily in mold fill thin sections

(d) Decrease in volume accompanying freezing of alloy and measure of amount of compensating feed metal

required in form of risers

(e) Based on resistance of alloy in standard salt spray test

(f) Composite rating, based on ease of cutting, chip characteristics, quality of finish, and tool life In the case of

heat-treatable alloys, rating is based on T6 temper Other tempers, particularly the annealed temper, may have lower ratings

(g) Based on ability of material to be fusion welded with filler rod of same alloy

Binary hypoeutectic alloys are too soft to have a good machinability rating However, the machinability of silicon alloys is generally very good in terms of surface finish and chip characteristics Tool life can be short with

aluminum-conventional carbide tools, particularly in the case of the hypereutectic alloys With the recent introduction of diamond cutting tools, tool life has been significantly increased, making the machining of the hypereutectic alloys practical

Corrosion resistance of these alloys is generally considered excellent Alloys containing increasing amounts of copper have a somewhat lower corrosion resistance than the copper-free alloys as measured in standard salt spray tests

Because of their high fluidity and good casting characteristics, these alloys are highly weldable with conventional welding techniques For joining purposes, brazing alloys and filler wire alloys (for example, alloys 4043 and 4047) (Ref 14) are also based on the aluminum-silicon alloy system

For wear applications, the important physical properties of these alloys include thermal expansion, thermal conductivity, electrical conductivity, and Young's modulus Data for these properties are available in standard references (Ref 13, 14,

15, 16, 17, 18) Because silicon is generally in precipitate form, the rule of mixtures is applicable when calculating the properties

Heat Treatment. Depending on the application, thermal treatments can be employed to:

In addition, aluminum-silicon and Al-Si-X alloys can be given a higher temperature (205 to 260 °C, or 400 to 500 °F)

aging treatment from the as-cast condition to improve their strength and thermal stability This is particularly important for applications where dimensional tolerances are critical (for example, when the alloy is operated at elevated temperatures as a piston component in an engine) Generally, such an aging practice is designated by the T5 temper

High-temperature (480 to 540 °C, or 900 to 1000 °F) treatments can also be given to aluminum-silicon and Al-Si-X alloys

to improve their ductility These thermal treatments modify the angular primary silicon particles to a more rounded shape This rounded shape reduces the tendency for crack initiation beginning at the sharp edges of the particles Such treatments are particularly effective on the hypereutectic alloys Other means of modifying the shape for improved ductility are discussed in the sections "Modification" and "Refinement" in this article

Principles of Microstructural Control. The three categories of aluminum-silicon alloys are based on the silicon level (Table 1) These alloy categories are hypoeutectic, eutectic, and hypereutectic The hypoeutectic alloys solidify with -aluminum as the primary phase followed by aluminum-silicon eutectic Other solutes (for example, iron, magnesium, and copper) form phases that separate in the freezing range of the alloy in the interdendritic locations (Fig 2)

Hypereutectic alloys solidify in a similar manner, but in these alloys silicon is the primary phase rather than -aluminum (Fig 3a) The eutectic alloys solidify principally with an aluminum-silicon eutectic structure; either aluminum or silicon

is present as a primary phase depending on which side of the eutectic composition (12.7% Si) the alloy lies A brief description of microstructural control is given below; additional information is available in Ref 2 and 12

Grain Structure. The grain size of the primary aluminum is controlled through the addition of heterogeneous nuclei to the melt in the form of master alloy inoculants such as Al-6Ti or Al-Ti-B (in the latter, the titanium can range from 3 to 5% and the Ti:B ratio from 3:1 to 25:1) Grain sizes vary from 100 to 500 m ( 0.004 to 0.020 in.) An example of the effect of grain refinement by an Al-Ti-B refiner is shown in Fig 4

Fig 4 Effect of grain refinement by the addition of an Al-5Ti-0.2B master alloy to type A356.0 (a) Without

titanium addition (b) With 0.04% Ti addition Etched with Poulton's reagent 0.85×

Cell Size. The interdendritic arm spacing (or cell size) is controlled by the cooling rate (Ref 19), which is in turn a function of the casting process and section thickness The smallest cell size is achieved with thin-wall high-pressure die casting At the other extreme, thick-wall and castings exhibit the largest dendrite cell size Casting processes such as low-pressure die casting and permanent mold casting provide intermediate solidification rates and consequently cell sizes that lie between the two extremes In a similar fashion to the cell size, the constituent phase size is largely controlled by the freezing rate

Modification. The term modification refers to the change in morphology and spacing of the aluminum-silicon eutectic phase induced by the addition of a chemical agent such as sodium or strontium There is a change from large divorced silicon particles to a fine coupled aluminum-silicon eutectic structure with an addition of approximately 0.001% Na or 0.005% Sr to the melt Varying degrees of modification (Fig 5) are obtained with lower levels of addition For details of the mechanism and practice of modification, see Granger and Elliott (Ref 12)

Fig 5 Variation in microstructure as a function of the degree of modification The modification level increases

from A to F; thus microstructure F is highly modified Source: Ref 2, 12

Antimony is also used to modify (more accurately, refine) the eutectic structure in hypoeutectic and eutectic alloys, particularly in Europe and Japan Like sodium and strontium, it increases the fluidity of the alloys and improves mechanical properties Furthermore, it is permanent, allowing melts to be more effectively degassed, which, in turn, provides sounder castings The great disadvantage of antimony is that it poisons (or negates) the effect of sodium and strontium, and it also creates a problem in recycling An additional serious drawback is the potential for the formation of stibine gas, which is highly toxic Unlike sodium and strontium, which can be used to effectively modify eutectic

structures over a wide range of freezing rates, antimony provides eutectic refinement only at the relatively high rates experienced in die castings and some thin-wall permanent mold castings

Refinement. In hypereutectic alloys, the primary phase is silicon In order to provide the desired small well-dispersed silicon particles, phosphorus is added to the level of about 0.1% P through the addition of a master alloy such as Cu-10P The phosphorus combines with aluminum to form aluminum phosphide, AlP, which provides effective nuclei for the silicon phase much the same as TiB2-type particles are effective nuclei for -aluminum (Fig 3b) However, phosphorus also negates the effectiveness of sodium and strontium It does so by combining with them to form phosphides, which do not modify the eutectic structure Similarly, sodium and strontium reduce the effectiveness of phosphorus additions by refining the primary silicon phase

Gas Porosity. Hydrogen porosity can be controlled by maintaining gas levels at 0.10 cm3/100 g This is not readily accomplished, particularly when modification of the melt is being sought with the addition of sodium or strontium However, gas fluxing methods are available (Ref 20) that provide the means of reducing hydrogen levels to the desired range

Also deleterious to casting soundness is the presence of nonmetallic inclusions that act as nuclei for gas pores Various molten metal filtration systems are available for inclusion removal (Ref 20)

Sludge. A problem experienced with aluminum-silicon alloys is the formation of hard intermetallic phases of the Al(FeM)Si-type, which settle out under gravity from the melt (Fig 6) The conditions that favor the formation of these phases are low holding temperatures (which are often employed in the die-casting industry); a quiescent melt; and relatively high levels of iron, manganese, and chromium The relative tendency to form sludge in the holding furnace is given by a segregation factor (SF):

silicon alloys because of the variety of microstructures that can be achieved as the alloys are processed for particular applications The relative effects of silicon particles, matrix hardness, and intermetallic constituents on the wear resistance

of aluminum-silicon alloys are summarized below

Silicon Particles. Under relatively light load conditions, which are normally associated with low (<10-11 m3/m) losses, wear resistance is not a strong function of silicon content (Ref 22, 23, 24, 25) In general, however, silicon additions to aluminum will increase the wear resistance The principal mechanism appears to be the influence of the hard silicon particles, which lead to higher overall levels of hardness The fact that the hard silicon particles are surrounded by a softer and relatively tough matrix improves the overall toughness of the material and can contribute to wear resistance by favoring more plastic behavior The eutectic and hypereutectic silicon alloys, with increased volume fractions of hard primary silicon particles relative to the hypoeutectic alloys, might be expected to have the best wear resistance of the

aluminum-silicon alloys Andrews et al (Ref 26, 27), for example, found that increasing the silicon content in

hypereutectic alloys reduced wear However, binary alloy data (Ref 22, 28) indicate that the hypereutectic alloys are not necessarily the most wear resistant Clarke and Sarkar (Ref 28) found that there was a relative minimum in wear for

binary aluminum-silicon alloys at about the eutectic level, as did Jasim et al (Ref 22), especially at applied pressures

<100 kPa (<15 psi) Clarke and Sarkar attribute the effects of silicon in part to its effect on metal transfer mechanisms between the pin and countersurface (Ref 29) There is also evidence for increased wear resistance with refinement of the silicon particle morphology (Ref 30, 31)

It is clear, therefore, that microstructure-based explanations are needed to account for the variation in wear rates with silicon content Moreover, there is a need to account for the reduction in strength that occurs with increased silicon content (Ref 32, 33) The complex effects of composition on wear behavior suggest that wear resistance depends on other material properties (for example, fracture toughness) (Ref 34) Thus lower fracture toughness at higher levels of silicon could lead to higher wear rates if larger pieces of debris are created during the wear process Variations in toughness and strength with composition might also account for the apparent ability of the near-eutectic compositions to have a greater load-bearing capability at a given wear rate than either higher or lower silicon levels

Matrix Hardness. Increased matrix hardness is typically achieved through the heat treatment response produced by copper and magnesium additions Most commercial applications of aluminum-silicon alloys, in fact, depend on the increased strength achieved by heat treatment The improved wear resistance of precipitation-strengthened material

compared to solid solution strengthened material under low wear conditions was also noted by Soderberg et al (Ref 35)

using aluminum alloy 6061, which is strengthened primarily by Mg2 Si precipitates This is also the strengthening mechanism in the heat-treatable magnesium-bearing aluminum-silicon alloys Although heat treatment has a beneficial effect (Ref 26, 27, 32), variations in matrix hardness may be less important than the effects of silicon content (Ref 27)

Intermetallic Constituents. In addition, there are important "other" hard phases present in commercial

aluminum-silicon alloys that provide enhanced wear resistance These constituents (for example, Al-Fe-Si, Al-Fe-Mn, Al-Ni,

Al-Ni-Fe, Al-Cu-Mg) have varying degrees of hardness (Ref 36, 37, 38) Despite the apparent scatter, these constituents are all much harder than the aluminum matrix Some examples of the hardness values of these intermetallic compounds are shown in Table 4

Table 4 Typical hardness values of selected intermetallic constituents of aluminum-silicon alloys

The addition of "hard" phases in the form of particles or fibers to reduce wear is also utilized to create metal-matrix composites (MMC) materials (Ref 39) These materials utilize hard intermetallic, cermet, or ceramic phases to provide the high hardness material for wear resistance Hornbogen (Ref 40) and Zum Gahr (Ref 41) have described in quantitative terms how the contribution of these hard phases to wear resistance can be modeled in terms of their volume fraction and morphology This composite approach has been effectively used to develop new piston materials (see the section "Metal-Matrix Composites" in this article)

Finally, the use of "softer" constituents (for example, graphite) should also be noted as an active area for development of wear-resistant aluminum-silicon MMC materials (Ref 32, 39, 42) In these materials, ranking may depend on whether volumetric wear rates (in units of m3/m) or seizure resistance is being considered The presence of the softer phase may lead to greater volumetric wear in some cases but greater resistance to seizure (higher load at seizure) in other cases

To summarize, the results of wear studies using aluminum-silicon alloys illustrate a variety of mechanisms The effect of variations in silicon particle morphology is often not clear cut, although heat treatment is beneficial to the sliding wear resistance Therefore, selection of an optimum microstructure is often difficult in practical situations where several wear types or mechanisms could occur In general, either eutectic or hypereutectic alloys offer the greatest wear resistance under a wide range of wear conditions Selection may then hinge on the dependence of in-service performance on other alloy characteristics or cost Overall, the aluminum-silicon alloy system provides a good basis for developing lightweight, strong, wear-resistant materials Examples of these applications will be discussed in the following sections

Aluminum-Silicon Alloy Applications

Aluminum-silicon alloys are used in a variety of automotive, aerospace, and consumer product applications

Automotive Components

Table 5 lists typical automotive components made from aluminum-silicon casting alloys (Ref 43) The eutectic or nearly eutectic alloys (for example, 332, 336, and 339) (Ref 44), are perhaps the most widely used Equivalent versions of these alloys are used for similar applications by European and Japanese automakers (Ref 45, 46, 47)

Table 5 Automotive engine applications of aluminum-silicon alloys

SAE alloy Type of

355.0 S, PM Pump bodies, cylinder heads

390.0 D Cylinder blocks, transmission pump and air compressor housings, small engine crankcase, air conditioner pistons

A390.0 S, PM Cylinder blocks, transmission pump and air compressor housing, small engine crankcase, air conditioner pistons Source: Ref 43

Pistons. Typical applications for aluminum-silicon alloys in the French automotive industry are shown in Table 6 (Ref 45) In addition to being cast, the A-S12UN (eutectic) alloy can also be forged (Ref 48) Similarly, AA 4032, somewhat

similar in composition to 336, is also widely used as a piston alloy (for example, for high-performance forged pistons) Hypereutectic alloys are also used for cast pistons, especially in diesel engines (Ref 45, 49) The potential benefit from composites that combine the strength reinforcement of ceramics with an aluminum-silicon alloy matrix has also been evaluated (Ref 45, 47, 48)

Table 6 Selected aluminum-silicon alloy applications in automobiles produced in France according to engine type and specific automobile manufacturer

Manufacturer

Engine type

Citroen Peugot Renault Talbot

A-S12UN A-S10UN(F)(a) A-S12UN A-S10.5UN

to 6 in.) surface finish, followed by controlled etching/polishing to leave silicon particles standing slightly above the alloy surface, was deemed necessary for optimum wear resistance

Efforts to simplify the 390-type technology by finding a more wear-resistant alloy for the cylinder or reducing the difficulties of finishing the bore have led to substitute alloys One approach has been the use of a lower silicon alloy containing more nickel and manganese (for example, the Australian-3HA alloy, with a nominal composition of Al-13.5Si-0.5Fe-0.45Mn-0.5Mg-2Ni) (Ref 61)

Continuing interest in the use of more highly wear-resistant materials in other engine-related parts has led to recent applications such as roller-type valve rocker arms (Ref 62) and valve lifters for the Toyota Lexus (Ref 63, 64) The rocker arm alloy used in the Mazda 929 is a nominal Al-10Si-2.7Cu-0.8Mg-0.45Mn alloy somewhat similar to the AA 383 alloy The valve lifter, on the other hand, is a strontium-modified Al-Si-Cu alloy designated 4T12 (composition, Al-10.5Si-4.5Cu-0.6Mg-0.2Mn)

Typical examples from a more detailed compilation of aluminum alloys used in wear-resistant applications in U.S autos are shown in Table 7

Table 7 Wear-resistant aluminum-silicon alloys used in automotive piston components produced for United States automotive manufacturers in 1978 to 1985 model years

Internal combustion engine components

Brake system components

Wheel cylinder pistons

Master cylinder pistons

Bearing Alloy Components. Aluminum alloys have been utilized for bearing applications for many years The many

uses range from diesel and internal combustion engines to a variety of tooling applications (for example, presses, lathes, and milling machines) (Ref 66) Important cast bearing alloys were based on aluminum-silicon or Al-Sn-Cu alloys,

whereas wrought bearing alloys have included the 8xxx types (for example, AA 8081 and AA 8020) (Ref 66, 67)

Compositions of various aluminum bearing alloys are listed in Table 8

Table 8 Nominal compositions of standard aluminum-silicon alloys used in bearing applications

Advanced Aluminum Bearing Alloys. The nominal compositions of improved bearing alloys with silicon additions are listed in Table 9

Table 9 Nominal compositions of advanced aluminum-silicon alloys used in bearing applications

H 4 0.5 0.1 0.1 6 0.3 Mn 13

Although a soft phase (for example, tin) is normally considered desirable for avoiding seizure, the compatibility of an 11Si-1Cu alloy was better than that of the traditional aluminum-tin alloy (SAE 783) (Ref 71) The silicon-copper alloy also had much better fatigue resistance The improved properties resulted in applications such as diesel crankshaft and connecting rod bearings Nevertheless, in line with the concerns expressed by Davies (Ref 72), the harder silicon-containing alloy was more sensitive to misalignment-induced seizure

Al-The addition of silicon to aluminum-tin alloys containing lower tin levels than that of the 783 alloy provided a compromise between the conformability of the soft-phase material and the benefits of the harder silicon phase for improved wear and fatigue resistance (Ref 73) This alloy could apparently be used without the common lead alloy overlays employed for seizure resistance The importance of fatigue resistance was also emphasized in the improved Al-

Sn-Si alloys reported by Ogita et al (Ref 74) As shown in Table 9, these alloys are somewhat similar to those of Fukuoka et al (Ref 73)

As another alternative to the lead- or tin-containing alloys, a graphite-containing Al-12Si alloy has been successfully evaluated for bearings (Ref 75) The use of a modified lead plus indium addition to an Al-11Si-Pb bearing alloy has also been reported (Ref 76)

Japanese concerns with the pollution and toxicity aspects of cadmium-containing alloys have led to improved zinc bearing alloys (Ref 77) These have also been improved with silicon additions The additional matrix wear between the silicon particles is believed to create lubricant reservoirs that enhance seizure resistance However, overlays (lead-tin alloys) are still required for best conformability

aluminum-Finally, the combined effect of silicon and refinement of the silicon- and lead-bearing phases by rapid solidification processing has resulted in an improved Al-6Pb-4Si bearing alloy (Ref 78) This alloy has grown in usage recently and is projected to be used in 78% of the cars built in the United States during 1991

Because of the increasing understanding of the balance among wear, fatigue, and seizure resistance of bearing materials, silicon alloys have been used to develop new and improved bearing materials Further improvements will undoubtedly be necessary as engine operating conditions evolve toward higher temperatures and operating speeds

Consumer Electronics Components

The growth of this market, which encompasses video cassette recorders (VCRs), video tape recorders (VTRs), digital audio tape (DAT) applications, and other devices (for example, personal computers), has created numerous opportunities for the use of lightweight, relatively corrosion-resistant and wear-resistant aluminum-silicon alloys VTR cylinders are specifically cited by various Japanese authors (Ref 79, 80, 81) because the eutectic-type silicon alloys have low coefficients of friction against the tape

Aerospace Components

A nonautomotive engine application of aluminum-silicon alloys is the use of the 390-type alloys in an aircraft engine (the Thunder engine) (Ref 82)

Breakthroughs in Aluminum-Silicon Wear-Resistant Materials

Metal-Matrix Composites. Composite pistons were recognized early as a potentially viable application of MMC technology While some composite approaches, especially for the severe operating conditions of diesel pistons, recommended the addition of specific metallic inserts to achieve improved performance (Ref 83), the bulk of development efforts have gone into the incorporation of ceramic fibers

The use of ceramic fiber aluminum-silicon MMC materials for pistons is described in a variety of publications (Ref 47,

48, 84, 85, 86, 87, 88) There are clear benefits to strength at elevated temperatures and reduction of the thermal expansion coefficient These materials appear to be especially applicable in critical areas such as the top piston rings and

top land (a high-temperature area) The castability of the aluminum-silicon alloys is a favorable factor in their use as matrices, particularly because squeeze casting is one of the preferred fabrication routes for composite pistons

The property improvements at elevated temperatures have encouraged ongoing development of the MMC technology for automotive engine applications, including engine blocks The ability of the MMC approach to allow selective strengthening of the cylinder region was taken advantage of by Honda engineers (Ref 89), who utilized composite reinforcement of alloy ADC12 (a Japanese alloy similar to 383)in the manufacture of a die cast engine block

Powder Metallurgy. The combined benefits of high silicon content and refined silicon particle size on wear resistance are strong driving forces behind the use of P/M techniques for making aluminum-silicon alloy parts The P/M approach has been of special benefit to the hypereutectic silicon alloys One example of this is the use of P/M A-S17U4 alloy to make cylinder liners (Ref 90, 91) The properties of these alloys exceed those of standard alloys (Ref 92) In addition, high levels of additional elements can be utilized to obtain good strength and wear-resistant properties at elevated temperatures (Ref 93, 94, 95)

The refined microstructures available from the P/M fabrication of hypereutectic alloys have a beneficial effect on fatigue characteristics as well This attribute has been utilized in the production of rotors and vanes for rotary automotive air conditioners (Ref 96) For this application, P/M alloys with high levels of iron or nickel are blended with P/M 2024-type alloys to create alloys containing 17 to 20% Si and 5 to 8% Fe or Ni

Spray casting, as exemplified by Osprey processing, has been shown to offer benefits similar to powder metallurgy (Ref 97, 98) This has the potential for even greater cost savings, which is an important factor if the aluminum industry is

to compete successfully in the automotive market In a comparison of the structure in an Al-20Si alloy, Kahl and Leupp (Ref 97) showed that there was practically no difference in silicon particle size between the P/M and Osprey processing routes Furthermore, they claimed the the fine and uniform silicon particle size resulted in improved wear behavior compared with that of conventionally produced material, although details of their test procedure are not know Earlier work at Delft University (Ref 99) compared the rate of mass loss of an Al-20Si-3Cu-1.3Mg alloy rubbing against cast iron at pressure level of 5.5 MPa (800 psi) In this test, the Osprey material showed better resistance to wear than either the ingot metallurgy (I/M) or P/M samples The reason for the better performance of the Osprey product compared with the P/M material was not clear, but it was hypothesized that the slightly larger silicon particles of the Osprey product helped reduce the fretting wear

Coatings/Surface Treatments. Other approaches to the wear (adhesion) problems of aluminum pistons moving in aluminum cylinders have taken the path of coating or surface treatment of the piston rings and cylinder bore to minimize wear problems (Ref 100) The selective fiber strengthening noted above is related to this problem also One widely used treatment is the so-called Nikasil treatment (Ref 101), an electrochemical treatment utilizing a dispersion of silicon carbide particles, preferably <4 m (<160 in.) in size, in a nickel matrix

Efforts to take advantage of both the P/M and surface treatment approach are exemplified by the emerging surface treatment technologies of surface alloying via ion implantation (Ref 102, 103, 104, 105), thermal spraying (Ref 106, 107, 108), and surface treating or alloying using laser treatments (Ref 109, 110, 111) Ion implantation is recognized for its ability to impart wear-resistant surfaces, but principal applications have been to protect tool surfaces in critical processing operations Laser hardening can be achieved through laser cladding to produce a chemically different surface or through

the effective heat treatment (or remelting and rapid solidification) brought about by laser heating The work by Blank et

al (Ref 111) using 7 and 12% Si alloys, however, indicated that surface alloying (for example, with iron or iron plus

vanadium) was more effective for increased wear than surface heat treatment effects alone In any case, the ability to tailor surface properties to a technological need will enable engineers to obtain further enhancement of the wear resistance

of the aluminum-silicon base alloys without sacrificing their other advantages

4 A Kaye and A Street, Die Casting Metallurgy, Butterworths Scientific, London, 1982, p 32

5 S Rabinowitz, The Expanding Role of Aluminum in the Automotive Industry, Die Cast Eng., Sept-Oct 1,

10 L.F Mondolfo,Aluminum Alloys: Structure and Properties, Butterworths, 1976

11 D.A Granger, R.R Sawtell, and M.M Kersker, Effect of Beryllium on the Properties of A357.0

Castings,Trans AFS, 1984, p 579-586

12 D.A Granger and R Elliott, Aluminum-Silicon Alloys, Casting, Vol 15, 9th ed., Metals Handbook, ASM

International, 1988, p 159-168

13 J.E Hatch, Aluminum: Properties and Physical Metallurgy, American Society for Metals, 1984, p 344-345

14 Source Book on Selection and Fabrication of Aluminum Alloys,American Society for Metals, 1978

15 A.L Kearney and E.L Aluminum Foundry Products, Properties and Selection: Nonferrous Alloys and

Special-purpose Materials, Vol 2, ASM Handbook, ASM International, 1990, p 123-151

16 A.L Kearney, properties of Cast Aluminum Alloys, Properties and Selection: Nonferrous Alloys and

Special-Purpose Materials, Vol 2, ASM Handbook, ASM International, 1990, p 152-177

17 J.E Hatch, Chapters 8 and 9, Aluminum: Properties and Physical Metallurgy, American Society for

Metals, 1984, p 152-177

18 General Information Chemical Compositions, Mechanical and Physical Properties of SAE Aluminum

Casting Alloys SAE J452, June 1983, SAE Handbook, Society of Automotive Engineers, 1989, p

10.5-10.13

19 R.E Spear and G.R Gardner, Dendrite Cell Size, Trans AFS, Vol 71, 1963, p 209-215

20 D.V Neff, Non-Ferrous Molten Metal Processes, Casting, Vol 15, 9th ed., Metals Handbook, ASM

International, 1988

21 J Gobrecht, Gravity-Segregation of Iron, Manganese and Chromium in an Aluminum-Silicon Casting

Alloy, Part I, Giesserei, Vol 62 (No 10), 1975, p 263-266; and D.A Granger, Investigation of a Star-Like Intermetallic Phase Occurring in Alloy 339, Trans AFS, Vol 91, 1991, p 379-383

22 K Mohammed Jasim and E.S Dwarakadasa, Wear in Al-Si Alloys under Dry Sliding Conditions, Wear,

Vol 119, 1987, p 119-130

23 B.N Pramila Bai and S.K Biswas, Effect of Load on Dry Sliding Wear of Aluminum-Silicon Alloys,

ASLE Trans., Vol 29 (No 1), 1986, p 116-120

24 R Shivanath, P.K Sengupta, and T.S Eyre, Wear of Aluminum-Silicon Alloys, Wear of Materials 1977,

International Conference on Wear of Materials, W Glaeser, K Ludema, and S Rhee, Ed., American Society of Mechanical Engineers, p 120-126

25 R Antoniou and D.W Borland, Mild Wear of Al-Si Binary Alloys during Unlubricated Sliding, Mater

Sci Eng., Vol 93, 1987, p 57-72

26 J.B Andrews, M.V Seneviratne, K.P Zier, and T.R Jett, The Influence of Silicon Content on the Wear

Characteristics of Hypereutectic Al-Si Alloys, Wear of Metals 1985, K Ludema, Ed., American Society

of Mechanical Engineers, p 180-184

27 J.B Andrews, M.V Seneviratne, and T.R Jett, The Influence of Si and Other Alloying Additions on the

Wear Characteristics of Hypereutectic Al-Si Alloys, 4th European Tribology Congress (EUROTRIB 85),

Vol III, 1985

28 J Clarke and A.D Sarkar, Wear Characterization of As-Cast Binary Aluminum-Silicon Alloys, Wear, Vol

54, 1979, p 7-16

29 J, Clarke and A.D Sarkar, The Role of Transfer and Back Transfer of Metals in the Wear of Binary Al-Si

Alloys, Wear, Vol 82, 1982, p 179-185

30 T.K.A Jaleel, N Raman, S.K Biswas, and K.K Murthy, Effect of Structural Modification and Load on

the Wear of a Hypereutectic Aluminum-Silicon Alloy, Aluminum, Vol 60 (No 12), 1984, p 932-933

31 Y Odani, K Akechi, and N Kuroishi, High Strength and High Wear Resistance Al-Si P/M Alloys,

Modern Developments in Powder Metallurgy, Vol 16, 1984, p 493-505

32 S Das, S.V Prasad, and T.R Ramachandran, Microstructure and Wear of Cast (Al-Si Alloy) Graphite

Composites, Wear, Vol 133, 1989, p 173-187

33 B.N Pramila Bai, S.K Biswas, and N.N Kumtekar, Scanning Electron Microscopy Study of Worn Al-Si

Alloy Surfaces, Wear, Vol 87, 1983, p 237-249

34 K.-H Zum Gahr, Microstructure and Wear of Materials, Tribology Series, Vol 10, Elsevier, 1987

35 S Soderberg, U Bryggman, and A Canales, Influence of Precipitation and Solution Strengthening on

Abrasive Wear Resistance, Wear of Metals 1985, K Ludema, Ed., American Society of Mechanical

Engineers, p 645-653

36 E.R Petty, Hot Hardness and Other Properties of Some Binary Intermetallic Compounds of Al, J Inst

Met., Vol 89, 1960-1961, p 343-349

37 A.A Ivanko and G.V Samsonov, Handbook of Hardness Data, Translation from Russian, National

Bureau of Standards-National Science Foundation, 1968

38 H.E.N Stone, The Oxidation Resistance, Hardness and Constitution of Metallic Aluminides, J Mater

Sci., Vol 10, 1975, p 923-934

39 S.V Prasad and P.K Rohatgi, Tribological Properties of Al Alloy Particle Composites, J Met., Nov 1987,

p 22-26

40 E Hornbogen, Description of Wear of Materials with Isotropic and Anisotropic Microstructure, Wear of

Metals 1985, K Ludema, Ed., American Society of Mechanical Engineers, p 477-484

41 K.-H Zum Gahr, Abrasive Wear of Two Phase Metallic Material with a Coarse Microstructure, Wear of

Metal 1985, K Ludema, Ed., American Society of Mechanical Engineers, p 45-57

42 B.N Pramila Bai, E.S Dwarakadasa, and S.K Biswas, Scanning Electron Microscopy Studies of Wear in

LM13 and LM13-Graphite Particulate Composite, Wear, Vol 76, 1982, p 221-220

43 General Information Chemical Compositions, Mechanical and Physical Properties of SAE Aluminum

Casting Alloys, SAE Handbook, J452, June 1983, Society of Automotive Engineers

44 R Wilson, Piston Improvement Relies on Material Science, Diesel Prog North Am., Vol 52 (No 2), Feb

1986, p 18-20

45 J Ayel, The Influence of Metallurgical and Geometric Parameters on the Wear of Recent Engines, Ing

Automob., Sept-Oct 1980, p 82-98

46 M Coulondre and J Travaille, New Trends in the Design and Manufacture of Pistons and Cylinders for

Automobile Engines, Ing Automob., Sept-Oct 1980, p 68-72

47 A.R Baker and S Gazzard, Developments in Materials for Pistons, Mater Des., Vol 9 (No 1), Jan-Feb

1988, p 28-33

48 J.C Avezou and A Parker, Evaluation of Materials for Pistons, Segments and Valve Seatings for Petrol

and Diesel Engines of Automobiles, Ing Automob., Oct 1983, p 73-80

49 J.P Whitacre, "Hypereutectic Aluminum A Piston Material for Modern High Specific Output Gasoline Engines," TP 871944, Society of Automotive Engineers, 1987

50 M Mazodier, Aluminum Engine Blocks with Thin Hypersilicon Alloy Liners, Ing Automob., No 10, Oct

1977, p 586-590

51 R El Haik, The Possibilities of Using Al Cylinder Blocks Hypersiliconized Cylinder Liner Tests, Ing

Automob., No 10, Oct 1977, p 569-577

52 J.L Jorstad, "Development of the Hypereutectic Al-Si Die Casting Alloy Used in the Vega Engine Block,"

TP No 105, 6th International Society of Die Casting Engineers Congress (Cleveland, OH), Nov 1970

53 J.L Jorstad, The Hypereutectic Aluminum-Silicon Alloy Used to Cast the Vega Engine Block, Mod Cast.,

Vol 60 (No 4), Oct 1971, p 59-64

54 J Ayel, Frictional Behavior of the Aluminum Hypersilicon Alloys Used as Engine Liners, Rev Alum., Apr

1978, p 186-191

55 H.H Hofmann and K Schellmann, "Aluminum 390 Alloy Engine Blocks: Design and Manufacturing," TP

830007, Society of Automotive Engineers, 1983

56 W Dahm and R.G Putter, "Light Alloy Engines Experience Gained from Research and Development,"

TP 830005, Society of Automotive Engineers, 1983

57 G Renninger, D Abendroth, and M Bolien, "Castings Engine Blocks in GK Al Si 17 Cu 4 Mg," TP

830003, Society of Automotive Engineers, 1983

58 E.G Jacobsen, "General Motors 390 Aluminum Alloy 60 degree V6 Engine," TP 830006, Society of Automotive Engineers, 1983

59 H.M Ward, III, "Engineering Control for High Volume 390 Die Casting," TP 830009, Society of Automotive Engineers, 1983

60 J.L Jorstad, "Reynolds 390 Engine Technology," TP 830010, Society of Automotive Engineers, 1983

61 J Barresi, R Giannini, and B.J Clarsen, "Cold Scuff Evaluation of 3HA The Engine Alloy Developed in Australia," TP 871203, Society of Automotive Engineers, 1987

62 S Komatsu, S Yahata, K Fukui, and T Tokoshima, "Development of Aluminum Alloy Roller Type Valve Rocker Arm by Casting and Forging Technique," TP 890554, Society of Automotive Engineers,

1989

63 S Ezaki, M Masuda, H Fujita, S Hayashi, Y Terashima, and K Motosugi, "Aluminum Valve Lifter for Toyota New V-8 Engine," TP 900450, Society of Automotive Engineers, 1990

64 H Baker, The Road Ahead for Metals in Autos, Adv Mater, Proc., May 1990, p 27-34

65 Aluminum Association Guide to Aluminum Parts 1978-1984, Reynolds Aluminum 1985 Aluminum Automotive Applications

66 Chapter 20, Aluminum Vol II: Design and Application, K.R Van Horn, Ed., American Society for Metals,

1967

67 Chapters 8 and 9, Aluminum Vol I: Properties, Physical Metallurgy and Phase Diagrams, K.R Van

Horn, Ed., American Society for Metals, 1967

68 "Bearing and Bushing Alloys," J460e, SAE Handbook, Society of Automotive Engineers

69 "Bearing and Bushing Alloys," J459c (Information Report), SAE Handbook, Society of Automotive

73 T Fukuoka, H Kato, and S Kamiya, Aluminum Alloy Bearings Containing Hard Particles Fitted for Use

with Nodular Iron Shaft, Studies of Engine Bearings and Lubrication, SP-539, Society of Automotive

Engineers, 1983, p 33-44 (TP 830308)

74 Y Ogita, K Niwa, and Y Kondo, "New Engine Type Fatigue Phenomenon of Al-Based Engine Bearings," TP 890556, Society of Automotive Engineers, 1989

75 B.P Krishnan, N Raman, K Narayanaswamy, and P.K Rohatgi, Performance of Aluminum Alloy

Graphite Bearings in a Diesel Engine, Tribol Int., Vol 16 (No 5), 1983, p 239-244

76 D.P Howe and J.D Williams, An Al-Si-Pb-In Plain Bearing Material, Proceedings Fourth Conference of

Irish Durability and Fracture Committee (IDFC4), 1986, p 183-191

77 M Sakamoto, Y Ogita, Y Sato, and T Tanaka, "Development of New Al-Zn-Si Bearings for Heavy Load Applications in Uprated Engines," TP 900124, Society of Automotive Engineers, 1990

78 Automot Eng., Vol 97 (No 12), 1989, p 21-25

79 S Kitaoka, H Kamio, and S Kobayashi, Material Properties of Al-Si Alloys for Forging and Their

Applications, translation from Alutopia (Jpn.), Vol 17 (No 10), 1987, p 9-18

80 S Kitaoka, C Fujikura, and A Kamio, Aluminum-Silicon Alloys, translation of J Jpn Inst Light Metals,

Vol 38 (No 7), 1988, p 426-446

81 O Takezoe and Y Yasuda, Technical trends of Highly Wear Resistant Aluminum Alloy, Kobe Res Dev.,

Vol 38 (No 4), Oct 1988, p 23-26

82 R.P MacCoon and R.P Ernst, "The Thunder Aluminum 390 Alloy Aircraft Engine," TP 830008, Society

of Automotive Engineers, 1983

83 Automot Eng (London), Vol 7 (No 3), June-July 1982, p 25-27

84 M.W Toaz and M.D Smalc, Clevite's Advanced Technology for the Diesel Piston, Diesel Prog North

Am., Vol 51 (No 6), June 1985, p 56

85 R.R Bowles, D.L Mancini, and M.W Toaz, Design of Advanced Diesel Engine Pistons Employing

Ceramic Fiber Reinforcement, Seminar Advanced Composites: The Latest Developments, Nov 1986,

American Society for Metals, p 21-27

86 T Donomoto, K Funatani, N Miura, and N Miyake, "Ceramic Fiber Reinforced Piston for High Performance Diesel Engines," TP 830252, Society of Automotive Engineers, 1983

87 T Suganuma and A Tanaka, Applications of Metal Matrix Composites to Diesel Engine Pistons, J Iron

Steel Inst Jpn., Vol 75 (No 9), Sept 1989, p 1790-1797

88 R Keribar, T Morel, and M.W Toaz, "An Investigation of Structural Effects of Fiber Matrix Reinforcement in Aluminum Diesel Pistons," TP 900536, Society of Automotive Engineers, 1990

89 T Hayashi, H Ushio, and M Ebisawa, "The Properties of Hybrid Fiber Reinforced Metal and Its Application for Engine Block," TP 890557, Society of Automotive Engineers, 1989

90 R Perrot, Engine Sleeves Made in a Hypereutectic Aluminum-Silicon Alloy by Powder Metallurgy,

European Symposium on Powder Metallurgy (Stockholm), Vol 2, 1978, p 212-217

91 R Perrot, Engine Cylinder Liners in High Silicon Aluminum Alloy Produced by Powder Metallurgy, Rev

Alum., Mar 1979, p 129-135

92 N Kuroishi, Y Odani, and Y Takeda, High Strength High Wear Resistance Aluminum-Silicon P/M

Alloys, Met Powder Rep., Vol 40 (No 11) Nov 1985, p 642-645

93 J.L Estrada and J Duszczyk, Characteristics of Rapidly Solidified Al-Si-X Powders for High Performance

Applications, J Mater, Sci., Vol 25 (No 2A), Feb 1990, p 886-904

94 K Hummert and V Arnhold, Possibilities of Powder Metallurgy Aluminum Materials in Motor

Construction, 3rd International Symposium on Aluminum + Automobile, 1988, p 171-175

95 "High Strength Aluminum Alloy Powder Having Heat and Wear Resistance: Contains Silicon, Nickel, Iron and Manganese, and is Used for Cylinder Liners," U.S patent 4,938,810

96 T Hayashi, Y Takeda, K Akechi, and T Fujiwara, "Rotary Car Air Conditioner Made with P/M Al-Si Wrought Alloys," TP900407, Society of Automotive Engineers, 1990

97 W Kahl and J Leupp, "High Performance Aluminum Produced by Spray Deposition," BNF 7th International Conference Material through the 90s

98 F Hehermann et al., "Comparison of Processing Atomized and Spray-Deposited Al-20 Si-5-Fe-2 Ni

(wt.%) Aimed at Automotive Applications," PM90, Institute of Metals, London

99 J Duszczyk et al., The Osprey Preform Process and the Application to Aluminum Alloys Including Wear and High Temperature Resistance Al-20 Si-Cu Alloys, Modern Developments in Powder Metallurgy, Vol

18-21, Metal Powder Industries Federation, 1988, p 441-453

100 D.A Parker, Material Innovation and Its Impact on the Development of Components for Reciprocating

Engines, Int J Veh Des., IAVD Congress on Vehicle Design and Components, 1985, Inter-science

Enterprises, Ltd., United Kingdom, p B35.1-30

101 A.E Ostermann, "Experiences with Nickel-Silicon Carbide Coatings in Cylinder Bores of Small Aluminum Engines," TP 790843, Society of Automotive Engineers, 1979

102 J.D Destefani, Ion Implantation Update, Adv Mater Proc., Oct 1988, p 39-43

103 C.M Preece and J.K Hirvonen, Ion Implantation Metallurgy, TMS-AIME, 1980

104 I.L Singer, Tribomechanical Properties of Ion Implanted Metals, Ion Implantation and Ion Beam

Processing of Materials, MRS Symposia Proceedings, Materials Research Society, Vol 27, 1983, p

585-595

105 J.H Hirvonen, Industrial Applications of Ion Implantation, Ion Implantation and Ion Beam Processing of

Materials, MRS Symposia Proceedings, Materials Research Society, Vol 27, 1983, p 621-629

106 G.L Kutner, Thermal Spray by Design, Adv Mater Proc., Oct 1988, p 63

107 M.L Thorpe, Thermal Spray Applications Grow, Adv Mater Proc., Oct 1988, p 69

108 E.J Kubel, Jr., Thermal Spraying Technology: From Art to Science, Adv Mater Proc., Dec 1987, p 69

109 C.W Draper, Laser Surface Alloying: The State of the Art J Met., June 1982, p 24

110 J.M Rigsbee, Surface Engineering by Laser and Physical Vapor Deposition Techniques, J Met Aug

1984, p 31

111 E Blank, M Pierantoni, and M Carrard, Laser Surface Modification of Al-Si Cast Alloys for Wear

Resistance and Surface Toughness, Surface Modification Technologies III, 1990, T.S Sudarshan and D.G

Bhat, Ed., TMS-AIME, 1990, p 493-510

This article will discuss the raw materials used in the production of cemented carbides; the manufacturing methods employed; their physical, mechanical, and thermal properties; and the wear mechanisms encountered in service Emphasis

is placed on tungsten carbide-cobalt (WC-Co) or tungsten carbide-nickel (WC-Ni) materials as used in nonmachining applications

Acknowledgements

The author gratefully acknowledges the assistance of R James Franz in preparation of the photomicrographs, and the critiques by Steve J Burden and Les J Kastura of GTE Valenite, and J Gary Baldoni and Steve T Wayne of GTE Labs, Inc

Table 1 Properties of refractory metal carbides and binder materials

GPa 10 6 psi

Thermal expansion,

m/m · K Carbide

Tungsten carbide (WC) is manufactured through the reduction of tungsten oxide and subsequent carburization at 1400

to 1500 °C (2550 to 2730 °F) Particle sizes range from 0.5 to 30 m Each particle is composed of numerous tungsten carbide crystals The tungsten powder is sometimes doped with small (<1 wt%) amounts of vanadium, chromium, or tantalum/niobium before carburization These materials act as grain-growth inhibitors, particularly in the very fine (<1 m) particle sizes

Cobalt (Co) is the most widely used binder in WC-base hardmetals Cobalt is the preferred binder due to its outstanding wetting and adhesion characteristics As shown in Table 1, cobalt has a low-temperature hexagonal phase and as high-temperature cubic phase, with a phase transition at about 415 °C (780 °F) Cobalt is manufactured through reduction of cobalt oxides or derived from organic salts particularly cobalt oxalate The cobalt binder phase is altered significantly during milling with WC and subsequent liquid-phase sintering operations

Nickel (Ni) is used as a binder in less than 10% of total carbide production because of poor WC wettability, which results in decreased hardness and toughness relative to cobalt grades at identical binder levels Tungsten carbide-nickel grades offer slightly improved corrosion and oxidation resistance over cobalt binder grades

Tantalum/titanium/niobium carbides (TaC/TiC/NbC) are used predominately in metal forming applications when metal pick up or galling of dies is problem Tantalum carbide, in particular, is used in small quantities (<1 wt%) as a grain-growth inhibitor, and in fairly large amounts (>6 wt%) to provide increased hot hardness in metal cutting

Chromium carbide is added to both WC-Co and WC-Ni grades is small quantities (<5 wt%) to improve corrosion and oxidation resistance The WC-Ni grades alloy with chromium or chromium carbide have shown significantly improved corrosion resistance over conventional WC-Co or WC-Ni grades Chromium or chromium carbide additions increase the tendency to from a brittle carbon-deficient phase, which can result in decreased strength and toughness

Manufacturing Methods

Grade Powders are produced by combining WC with cobalt or nickel binder and, depending on the application, varying amounts of TaC/TiC/NbC The grade powder is milled in conventional ball mills, attritor mills, or vibratory mills The milling process reduces the particle size of the raw materials and also provides uniformity of mixture Milling operations are typically carried out in a protective solvent such as alcohol, used to minimize heating and subsequent oxidation of the powder, and to disperse the powder particles to achieve intimate mixing During the grade powder manufacturing process, 2 to 3 wt% of a solid lubricant such as paraffin wax is added This lubricant reduces the potential for oxidation and provides green strength to as-pressed components The lubricant/solvent/powder slurry is then dried to remove the solvent Spray drying is the most widely used atomized through a nozzle and sprayed into a stream of nitrogen gas The solvent is vaporized, condensed outside the chamber, and reused The dried powder is now in the form of free-flowing spherical aggregates on the order of 150 to 250 m in diameter (Fig 1)

Fig 1 Spray-dried cemented carbide powder 40×

Pressing or powder consolidation processes include a number of vastly different techniques Many wear components are produced on automatic or semiautomatic presses at pressures of 50 to 150 MPa (7 to 22 ksi) Press tooling typically consists of carbide dies, punches, and core rods to minimize wear In cold isostatic pressing, the grade powder is consolidated into a rough billet or ingot using equal pressure from all directions This rough billet is then preformed or machined to a net shape Extrusion of carbide grade powders is used to produce long components of small, constant cross section Higher lubricant levels ad different lubricant types are used in extrusion There has been limited injection molding of cemented carbides

Preforming or shaping operations are used to machine components to a net shape using abrasive (diamond) grinding wheels or single-point diamond tools These operations are used when the final part cannot be pressed to its final shape, or the production quantity is too low to justify te investment in press tooling

Sintering operations are carried out in batch-type or semicontinuous furnaces in either a vacuum hydrogen, or other inert atmosphere A 400 to 500 °C (750 to 930 °F) hold dewaxes the parts, and the vaporized lubricant is condensed outside the heating chamber and discarded Final sintering takes place at 1300 to 1600 °C (2370 to 2910 °F); the precise temperature depends on the cobalt content the grades with the higher cobalt contents have the lower sintering temperatures This final sintering temperature is above the eutectic temperature of the carbide-binder system, and the binder partially melts The excellent wettability of WC by cobalt results in rapid liquid-phase sintering, which promotes coalescence of the WC particles and produces a fully dense, virtually porosity-free microstructure Linear shrinkage on the order of 15 to 25% takes place

The advantages of hot isostatic pressing (HIP) have been exploited since the early 1970s Components requiring high reliability and/or surface integrity are HIPed to eliminate residual porosity, pits, or flaws Materials are heated to a temperature above the liquidus, and the vessel is pressurized with an inert gas to sightly less than 100 MPa (15 ksi) The combination of pressure and temperature forces the binder into any residual pits or porosity The result is porefree component with higher reliability Recent advances have combined liquid-phase sintering and HIP into a single sinter-HIP process Sinter-HIP uses lower pressures and higher temperatures than conventional HIP with no sacrifice in component reliability The resulting sinter-HIP microstructure is more uniform than that produced by conventional HIP, and sinter-HIP is more cost effective (Ref 1)

Finishing operations include grinding with diamond wheels, electrical discharge machining (EDM) using wire or shaped electrode, and edge honing using a variety of abrasive techniques Final lapping to mirrorlike finishes is accomplished using diamond-containing slurries or pastes

Physical or chemical vapor deposition (PVD or CVD) are now used on the majority of metal cutting inserts and

on some wear parts The coatings are on the order of 5 m thick and consist of titanium carbide (TiC), titanium nitride (TiN), aluminum oxide (Al2O3), or a combination thereof Their purpose is to minimize the wear process during steel machining; in particular, to minimize the dissolution of the workpiece material into the cutting tool Titanium nitride coatings are also purported to reduce the frictional forces at the tool/workpiece interface

The methods of application for CVD and PVD coatings differ substantially The CVD coatings are applied at temperatures of about 1000 °C (1830 °F), and as a result contain higher levels of residual stresses due to the difference in both thermal expansion coefficients and elastic moduli between the substrate and the coating This results in a significant decrease (30%) in three-point bending strength of coated parts compared to uncoated parts The PVD process is applied at less than 500 °C (930 °F) and results in reduced stresses in the coating and minimizes any loss in strength Coatings applied by PVD or CVD do not have the same utility in nonmachining applications where high-temperature wear resistance is not required Coatings may provide some improvement in corrosion resistance or resistance against smearing

or pickup of the workpiece onto a forming tool

Properties

The properties of cemented carbide grades are predominately determined by their chemical composition and the grain size

of the tungsten carbide in the sintered part Table 2 summarizes the properties of some typical carbide grades used in wear part application Table 3 relates these compositions to application areas

Table 2 Properties of representative cobalt-bonded cemented carbide grades

MPa ksi MPa ksi GPa 10 6

psi

Relative abrasion resistance(a)

at

200

°C (390

°F)

at

1000

°C (1830

°F)

Thermal conductivity,

(a) Based on a value of 100 for the most abrasion-resistant material

Table 3 Nominal composition and properties of representative cemented carbide grades and their

Heavy blanking punches and dies, cold heading dies 20-30 Medium 85

Heading dies (severe impact), hot forming dies, swaging dies 11-25 Medium to

coarse

84

Back extrusion punches, hot forming punches 11-15 Medium 88

Back extrusion punches, blanking punches and dies for high shear strength

Extrusion dies (medium impact), blanking dies, slitters 12-16 Medium 88

Corrosion-resistant grades, valves and nozzles, rotary seals, bearings 6-12 Fine to

medium

92

Corrosion-resistant grade with good impact resistance for valves and nozzles,

rotary seals and bearings

Deep draw dies (nongalling), tube sizing mandrels 10 Co with TiC and

TaC

Hardness is typically measured on the Rockwell A scale with values ranging from 83.0 HRA for high-cobalt grain grades, to 93.0 HRA for low-cobalt fine-grain grades Vickers diamond pyramid hardness (HV) is widely used in Europe; values range from 800 to 2000 kg/mm2 using a 30 kg load The precision and accuracy of hardness testing is influenced significantly by the surface finish of the testpiece, parallelism between top and bottom surfaces, and the quality

coarse-of hardness standards and diamond penetrators For straight WC-Co grades (those not containing TiC, TaC, or similar additions) with comparable WC grain size, hardness decreases with increasing binder content Figure 2(a) illustrates the relationship between hardness and cobalt content/WC grain size

Fig 2 Variation of properties with cobalt content and grain size for unalloyed grades of cemented carbide

Fracture toughness (KIc) values indicate the resistance of a material to fracture from intrinsic flaws A variety of test methods and specimen geometries are used, so caution must be exercised when comparing reported values from different manufacturers Fracture toughness increases with both increased cobalt content or WC grain size, as shown in Fig 2(b)

Density varies inversely with cobalt content, as shown in Fig 2(c) Porosity levels also influence density Cemented carbide grades used in ferrous alloy machining applications contain higher amounts of TiC/TaC and have density values from 10 to 14 g/cm3

Transverse rupture strength (TRS), or three-point bending strength, is the most common method of determining the fracture strength of cemented carbides Rectangular samples(5 × 6 × 19 mm, or 0.2 × 0.25 × 0.75 in.) are loaded as shown

in Fig 3 The TRS is then calculated using:

where F is the load at fracture, L is the span between supports, and W and H are the width and height of the test bar,

respectively The test itself is sensitive to test bar size (several variations are in use), surface finish, and other test parameters The response of TRS to cobalt content and tungsten carbide grain size is shown in Fig 2(d)

Fig 3 Schematic of transverse rupture strength testing (three-point bending)

Compressive strength of cemented carbides is greater than that of almost any group of materials, metallic or

nonmetallic Uniaxial compressive strength tests are performed using straight cylindrical samples, a cylinder with reduced diameter at the center of the part to localize fracture, or a straight cylinder held within a sleeve Reported compressive strength values can vary significantly depending on the size and geometry of the test specimen Compressive strength varies inversely with cobalt content, and for a given cobalt content, finer grain sizes give the highest value Figure 2(e) summarizes this compressive strength versus cobalt content/grain size relationship

Modulus of elasticity, or Young's modulus, also varies inversely with cobalt content (Fig 2f), but is independent of

WC grain size (Ref 2) The elastic modulus of WC is higher than that of any other commercially available material except diamond and cubic boron nitride As a result, WC-Co alloys have elastic moduli 2 to 3 times those of cast irons or steels

Elastic modulus measurements are carried out using either mechanical or sonic methods Mechanical methods involve loading a WC-Co beam and measuring the amount of deflection Sonic methods are more widely used and utilize resonant vibration of a cylindrical or square rod The resonant frequency depends on the dimensions of the testpiece, the density of the material, and the elastic modulus of the material

Thermal conductivity of WC-Co alloys is important in machining applications because the ability of the tool to conduct heat away from the tool/workpiece interface has a definite effect on tool performance In nonmachining applications, such as a rotary mechanical-pump seal, the tungsten carbide seal ring must have high thermal conductivity to ensure heat flow away from the rotary seal/stationary seal interface Thermal conductivity decreases with increasing cobalt content and is unaffected by WC grain size (Ref 3), as illustrated in Fig 2(g) The additional of TiC reduces the thermal conductivity significantly

Coefficients of thermal expansion are an important design consideration when using WC-Co materials The linear coefficient of thermal expansion of WC-Co increases with increasing cobalt content (Fig 2h), and is independent of grain size Typical low-carbon steels, tool steels, and stainless have thermal expansion coefficients 2 to 3 times greater than those of carbides In metal forming applications at elevated temperatures, such as warm forming or extrusion, this difference, must be taken into consideration when designing steel/carbide assemblies This expansion coefficient mismatch also complicates brazing operations when joining cemented carbides to metals

Porosity in carbides is typically less than 0.1 vol% ASTM procedure B 276 classifies porosity into three types:

• A-type: pores less than 10 m in diameter

• B-type: pores between 10 and 25 m in diameter

• C-type: porosity caused by the presence of uncombined carbon

Porosity is evaluated by observing a polished, unetched surface, and comparing it with standards supplied in the ASTM procedure

Microstructures for nine typical WC-Co grades are shown in Fig 4 The microstructures are characterized by the angular, sharp WC grains surrounded by the cobalt binder Tungsten carbide grain sizes as shown in Fig 4 vary from <1

m in the fine grain sizes to 8 to 10 m in the coarse grain size grades Microstructures of grades for use as cutting tools and anomalies such as graphite and the carbon-deficient phase are illustrated in the article "Cemented Carbides" in

Properties and Selection: Nonferrous Alloys and Special-Purpose Materials, Volume 2 of ASM Handbook (1990)

Fig 4 Cemented carbide microstructures (a) 94WC-6Co, fine grain size (b) 94WC-6Co, medium grain size (c)

93WC-7Co, coarse grain size (d) 10Co, fine grain size (e) 10Co, medium grain size (f) 10Co, coarse grain size (g) 84WC-16Co, fine grain size (h) 84WC-16Co, medium grain size (i) 75WC-25Co, medium grain size All at 1500×, 3 min etch

90WC-Wear Properties of Cemented Carbides

Superior abrasive wear resistance is probably the major reason for the selection of cemented carbides in a wide

variety of industrial applications This superior wear resistance can generally be attributed to their unique composition, which consists of 80 to 95% hard, wear-resistant, fine WC grains combined with a cobalt binder that provides a small amount of ductility to the material Studies of carbides have shown that abrasive wear involves rounding, fragmentation, and pullout of the WC grains, and subsequent removal of the soft binder phase (Ref 4, 5)

ASTM test B 611 specifies a method for determining the abrasive wear resistance of carbides In this test, a rectangular carbide testpiece is held against a rotating wheel (either steel or rubber) for a fixed number of revolutions The test is run either wet or dry A schematic of the test apparatus is shown in Fig 5 A steady stream of 30 mesh alumina sand is introduced directly into the carbide/wheel interface The volume of the testpiece (measured in cubic centimeters) is determined both before the test, and after the predetermined number of revolutions The volume loss is inversely related

to the wear resistance Typical volume losses for low-cobalt, fine-grain materials is on the order of 3 mm3 A D2 tool steel

is often used as reference material in this test and exhibits a volume loss of 40 to 45 mm3 Materials are ranked using either the reciprocal of the volume loss (low volume loss equates to high wear resistance), or using a particular grade (usually a WC-6Co grade with hardness of 92.0 HRA) as a reference, and reporting the performance of the grade as a percentage of this standard Some caution must be exercised when comparing values reported by different manufacturers Cemented carbide manufacturers have not agreed to a single test method, and there will be variation in the values and the units used to report wear resistance

Fig 5 Schematic of abrasive wear resistance apparatus

Another method involves a pin-on-disk apparatus (Ref 4) In this test, a resin-bonded diamond-covered disk is rotated at approximately 40 rev/min A load is applied through a square WC-Co sample that is swept and rotated against the diamond wheel The volume of the sample is determined before and after, and the volume loss is determined Values are reported in a manner similar to that of the ASTM procedure above

Figure 6 shows the relationship among fracture toughness, hardness, WC grain size, and abrasive wear resistance (Ref 4) The grades shown as 6C, 6M, 6F, 12C, 12M, and 12F represent 6 and 12% cobalt grades with coarse, medium, and fine grain size, respectively The abrasive wear resistance increase with decreasing cobalt content and decreasing grain size The abrasive wear resistance appears to be influenced more strongly by the WC grain size than by the cobalt content

Fig 6 Results of abrasive wear resistance tests See text for description of data points Source: Ref 4

The results of an abrasion test cannot be used to described the performance of a carbide grade when anything other than pure abrasive wear is present For example, when cratering or dissolution wear is taking place during machining

operations, abrasion resistance test results are not meaningful The same can be applied to metal pickup or galling during metal forming In summary, abrasive wear resistance test results can only be applied to those instances when abrasion is the dominant wear mechanism

Erosion resistance of carbides is important in applications such as sand blast-spray nozzles, seals in slurry pumps, and component parts in the oil industry The success of carbides as an erosion-resistant material is again due to its unique composite structure of wear-resistant WC particles in a ductile cobalt matrix (Ref 4, 6)

The erosion resistance test involves impacting the WC-Co surface with 50 m alumina particles at over 100 m/s (325 ft/s) for a set period of time under an inert gas carrier The volume of the resultant crater is determined and reported as a volume loss The smaller crater volumes equate to improved erosion resistance Figure 7 illustrates a test apparatus for gas stream alumina particle erosion tests

Fig 7 Schematic of specimen fixturing for particle erosion testing Source: Ref 6

The performance of carbides in erosion tests closely parallels their abrasive wear performance: the erosion resistance increases with decreasing cobalt content The effect of WC grain size is not as strong as in straight abrasion, but the erosion rate does decrease slightly with increasing carbide grain size for cobalt percentages greater than 15% Figure 8 shows the relationship between crater volume and the cobalt volume fraction This relationship suggests that a binderless carbide (<5 vol%) would be preferred for erosion applications However, grades with less than 5 vol% binder tend to exhibit higher wear rates due to the presence of increased grain boundary porosity after consolidation (Ref 6) These binderless alloys may not have the fracture toughness required to handle the thermal or mechanical loading that will be encountered in service As such, design engineers may be required to compromise some erosion resistance for improved structural integrity and mechanical reliability

Fig 8 Results of particle erosion tests Source: Ref 6

Corrosion resistance is not typically thought of as a requirement for carbides, and in terms of straight corrosion resistance, there are commercially available materials that are clearly superior However, when a combination of corrosion resistance, wear resistance, stiffness, toughness, and thermal conductivity is required such as for rotary mechanical seals in industrial pumps cemented carbides are the material of choice

The corrosion resistance of carbides is limited by the susceptibility of the cobalt binder to chemical attack, although there are some corrosive media that attack WC The corrosive media typically dissolve the cobalt binder from the matrix, leaving behind a weak, unsupported skeleton of tungsten carbide grains, which are easily abraded away The corrosion resistance of straight WC-Co alloys is, in general, inversely related to that of the binder content The straight substitution

of nickel binder for cobalt does provide limited improvement in both corrosion and oxidation resistance There are several grades available that utilize chromium additions to the cobalt binder to improve corrosion resistance Chromium, however, promotes the formation of the carbon-deficient phase with a resultant decrease in toughness and strength The development of nickel binder grades alloyed with chromium, molybdenum, and other elements has resulted in a significant improvement in corrosion resistance with little sacrifice in strength or toughness Figure 9 shows corrosion resistance versus pH for three different types of materials: WC-Co, alloyed WC-Ni, and TiC-Ni cermets (Ref 7) Although TiC-base cermets exhibit superior corrosion resistance, their use has been severely limited due to their inferior strength and lower thermal conductivity

Fig 9 Corrosion resistance of cemented carbides Source: Ref 7

Corrosion resistance is determined using immersion methods Test specimens are immersed in the media for a set period

of time, at a constant temperature, and at a known pH level The corrosion behavior of the material is determined by one

of three methods The first is a metallographic examination in which a cross section of the material is polished, etched, and observed at high magnifications to determine the depth of binder removal The second involves tumbling the test specimen after immersion Tumbling abrades the unsupported network of WC grains and leads to a final weight loss measurement The third method utilizes atomic absorption techniques to measure the amount of binder (cobalt or nickel) that has been dissolved in the corrosive media A low concentration of binder in the media indicates a high corrosion resistance Additional information on the corrosive behavior of carbides can be found in the article "Corrosion of

Cemented Carbides" in Corrosion, Volume 13 of ASM Handbook (1987)

References

1 R.C Leuth, Moldless Hot Pressing of Cemented Carbides, Carbide Tool J., Nov-Dec 1984, p 20-25

2 H Doi, Elastic and Plastic Properties of WC-Co Composite Alloys, Freund Publishing (Israel), 1974, p

32-40

3 A Perecherla and W Williams, Room Temperature Thermal Conductivity of Cemented Transition Metal

Carbides, J Am Ceram Soc., Vol 12, Dec 1988, p 1130-1133

4 S.F Wayne, J.G Baldoni, and S.T Buljan, Abrasion and Erosion of WC-Co with controlled Microstructures,

Tribol Trans., Vol 33 (No 4), 1990, p 611-617

5 J Larson-Basse, Binder Extrusion in Sliding Wear of WC-Co Alloys, Wear, Vol 105, 1985, p 247-256

6 D.K Shetty, J.T Stropki, and I.G Wright, "Slurry Erosion of WC-Co Cermets and Ceramics," Preprint No 84-AM-3A-3, American Society of Lubrication Engineers (ASLE), May 1984

7 L Lindholm, U.S patent 4,497,660, 5 Feb 1985

Selected References

• ASM Committee on Tooling Materials, Superhard Tool Materials, Tool and Die Failure Source

Book, American Society for Metals, 1982, p 400-410

• K.J.A Brookes, World Directory and Handbook of Hardmetals, 4th ed., International Carbide Data,

• P Schwarzkopf and R Kieffer, Cemented Carbides, Macmillan, 1960

Friction and Wear of Metal-Matrix Composites

P.K Rohatgi, Y Liu*, and S Ray**, University of Wisconsin-Milwaukee

Introduction

FOR MOST OF THE PAST CENTURY, research and development in tribology have been directed at meeting the severe conditions of mechanical systems, such as advanced engines, that require increasingly high working temperatures and running speeds The development of new materials that can meet these severe requirements has become a major imperative Metal-matrix composites are attracting considerable interest worldwide because of their superior mechanical and tribological properties These advanced materials have a metal matrix in which nonmetallic fibers, particles, or whiskers are dispersed For use in tribological applications, metal-matrix composites must be able to support a load without undue distortion, deformation, or fracture during performance and to maintain controlled friction and wear over long periods without seizure under working conditions (Ref 1)

This article will review the structure, processing, and properties of those metal-matrix composites (both particulate and fiber reinforced) that have been studied for their tribological properties In addition, it will discuss the current and potential uses of metal-matrix composites in applications where tribological properties are of paramount importance, such

as electrical contact brushes, bearings, pistons, cylinder liners, and brake drums

A variety of nonmetallic particles have been dispersed in different metal systems to develop metal-matrix nonmetallic particle composites; a selected list of these particles is given in Table 1 (Ref 2, 3, 4) These particles can be roughly divided into two different groups on the basis of hardness: (1) hard particles with hardness of 4 to 30 GPa (580 to 4350 ksi), such as SiC, Al2O3, and silica, and (2) soft particles with hardness below 2 GPa (290 ksi), such as graphite and MoS2, which are primarily added for solid lubrication purposes Particles with values between these ranges are considered

to have intermediate hardness This arbitrary division is based on the fact that composites containing hard particles generally exhibit different friction and wear behavior (because of different wear mechanisms) than composites containing soft particles in alloys with the same matrix

Table 1 Density and hardness of selected particles for metal-matrix particle composites

Table 2 lists a number of metal-matrix composites synthesized using the two types of particles, along with particle size, volume fraction, matrix alloys, and methods of fabrication The matrix alloys used for the synthesis of composites include those based on aluminum, copper, silver, iron, magnesium, steel, titanium, and cobalt Synthesis techniques include powder metallurgy, mechanical alloying, spray dispersion, squeeze casting, stir casting, compocasting, and low-pressure infiltration In recent years, the use of inexpensive liquid metallurgy techniques, including stir casting, squeeze casting, and pressure infiltration, has become widespread in the production of metal-matrix particulate composites (Ref 5, 6, 7)

Table 2 Selected metal-matrix nonmetallic particle composites and methods used for their synthesis

Particle type Dispersoid

size, m

Vol% Matrix alloy Method of fabrication

0.3-20

Al-Si, Al-Cu, Al-Cu-Mg Vacuum slurry casting, squeeze casting, powder metallurgy

TiC <40-212 8-40 Al-Cu, Al-Mg, Ti-Al-V,

Al-Mg, Al-Cu, Al-Si,

Cu, steel, Mg

Slurry casting, squeeze casting, powder metallurgy

WC 106-150 Ti-Al-V, Co-base Laser melt-particle injection, powder sintering

M 7 C 3 (Cr-rich) 18-38 Co-Cr Powder metallurgy

ZrO 2 /ZrSiO 4 5-80 1-4 Cu, Al, steel Slurry casting, bottom pouring, spray dispersion, powder

metallurgy

TiO 2 /MgO 10 Cu, Al, steel Slurry casting, bottom pouring, spray dispersion, powder

metallurgy

Glass/SiO 2 30-110 2-10 Al-Mg, Cu Slurry casting, bottom pouring, powder metallurgy

Mica/talc 40-180 3-10 Al-Cu-Mg, Ag, Cu-Sn Slurry casting, compocasting, powder metallurgy

Shell char 125 15 Al-Si-Mg Slurry casting, squeeze casting

Graphite 15-800 1-750 Al, Cu, Ag, iron Slurry casting, squeeze casting, powder metallurgy

PTFE 20-40 Cu, Ag, Cu-steel Powder metallurgy

MoSe 2 20-80 Fe-Pb, Ag-Cu, Ag Powder metallurgy

Notes

* Presently with National University of Singapore

** On leave from Department of Metallurgical Engineering, University of Roorkee, India

Mechanical Properties of Metal-Matrix Composites

The mechanical properties of a material are important in determining its tribological behavior Wear can often be attributed to cracks that originate in the heavily deformed subsurface region and propagate to the surface, thereby releasing debris Therefore, the deformation and fracture behavior of a material under both uniaxial and triaxial load assumes importance in the context of its tribological response

In metal-matrix composites, mechanical properties depend on the amount, size, shape, and distribution of the dispersed phase apart from the mechanical properties of the matrix material and on the nature of the interface By definition, a composite material generally requires an amount of dispersed phase (> 1 vol%) of a size (> 1 m) that allows this constituent to be load bearing and not act merely to control the movement of dislocations, as in dispersion-strengthened materials The shape of the dispersed phase is so important in determining its load-bearing capacity that composites have been classified on this basis: (1) fiber-reinforced composites with both continuous and discontinuous fibers and (2) particle- or whisker-reinforced composites The aspect ratio (that is, the ratio of length to diameter of a fiber) generally characterizes the shape In continuous-fiber composites, the load is applied directly to both the matrix and the fiber In discontinuous-fiber composites or particle-reinforced composites, the load is transmitted to the dispersoid through the matrix Mismatch of strain in the matrix and the dispersoid across the interface results in a shear stress at the interface The transfer of load to the dispersed phase depends on the magnitude of the shear stress as described in simplistic "shear lag" models for discontinuous fibers and particles in a matrix (Ref 8, 9) If the shear stress that develops at the interface exceeds the strength of the interface, debonding will occur at the interface, and this may develop into a crack that can propagate For a given condition of loading, the load shared by the dispersed phase increases with its aspect ratio

Going beyond the framework of elasticity theory, the presence of hard dispersed particles will cause additional strain hardening; this may not be significant in the case of soft particles with a shear modulus lower than that of the matrix alloy Thus, composites with soft particles such as graphite have lower strength compared with the matrix alloy, as shown

in Fig 1 (Ref 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25) The strength of composites that contain hard particles increases with the volume percentage of particles in the composite However, the ductility of composites that contain both soft and hard particles decreases compared with that of the matrix alloy (Fig 2) presumably because of debonding of the interface at low strain, as often indicated by serrations in tensile stress-strain diagrams (Ref 26)

Fig 1 Ultimate tensile strength as a function of particle volume fraction for aluminum alloy composites

Fig 2 Elongation as a function of particle volume fraction for aluminum alloy composites

The mechanical properties of discontinuous-fiber composites are similar to those of particle- or whisker-reinforced composites, except that the strength of discontinuous-fiber composites increases with aspect ratio In the case of continuous-fiber composites, strength increases with volume fraction of the fibers (Fig 3), even for soft fibers such as graphite (Ref 27), because these fibers are strong The physical properties of the fibers used to reinforce metal-matrix composites are given in Table 3 The mechanical properties of fiber-reinforced metal-matrix composites are given in Table 4

Table 3 Properties of fibers used in metal-matrix composites tested for tribological applications

Fiber Composition Diameter, m

g/cm 3 lb/in. 3 GPa ksi GPa ksi Thornel 50 Carbon 6 1.67 0.060 2.40 348 413 59,900

Saffil RF/Saflmax -Al2O3 3 3.30 0.120 2.00 290 300 43,500

SN-N-X (whisker) Si 3 N 4 0.2 3.20 0.116 3.00 435 350 50,760

Tokamax (whisker) SiC 0.5 3.20 0.116 10.00 1450 480 69,600

SiC whisker -SiC 0.1-1 3-10 435-1450 400-700 58,000-101,500

Thornel 300 38 6.19 0.224 Powder metallurgy

Al-4.7Cu (A2010) Thornel 50 30 Liquid infiltration

Al-4.7Cu (A2010) Thornel 300 30 Liquid infiltration

Al-1Mg (6061) Thornel 50 30 Liquid infiltration

Copper Thornel 50 30 Liquid infiltration

Tin Thornel 50 30 Liquid infiltration

Al alloy (ADC 12) SiC whisker 20-40 Squeeze casting

Cu-Zr PAN type (carbon)(a) 3-50 Impregnation

Ag-28/20Cu Thornel 300 42/44 6.58/6.41 0.238/0.232 Liquid infiltration

Ag-28Cu Celion 6000 40 6.80 0.246 Liquid infiltration

Ag-26Cu-2Ni Celion 6000 28 7.65 0.276 Liquid infiltration

(a) PAN, polyacrylonitrile

Fig 3 Ultimate tensile strength of squeeze-cast continuous graphite-fiber-reinforced aluminum alloy

composites as a function of fiber volume fraction

The fracture toughness of particulate composites is low, and it decreases with volume fraction of a dispersed phase such

as SiC, as shown in Fig 4 (Ref 10, 23, 28, 29) A weak interface leads to cracks at low strain, and a low fracture

toughness value indicates their easy propagation In the context of wear, these two properties are critical to the generation and propagation of cracks, resulting in wear debris Pores in composites may act as preexisting crack nuclei in the system, waiting to become unstable at appropriate stress levels (Ref 30) Porosity thus increases composite wear rate

Fig 4 Fracture toughness of aluminum alloy composites containing SiC as a function of volume fraction of SiC

During tribological interaction between two surfaces, temperature rises significantly at and near the surface, and the mechanical properties of the composites at elevated temperatures may be of direct relevance The dispersed hard particles are helpful in retaining the high-temperature strength of the matrix (Ref 31) Also, the loss of strength due to dynamic recrystallization is counteracted by the presence of particles Metal-matrix composites can thus have superior mechanical properties at elevated temperatures, even though the room-temperature properties may not look encouraging in certain systems

Tribological Behavior of Metal-Matrix Particulate Composites

Friction and Sliding Wear. Friction and wear behavior of metal-matrix composites depends on the nature of particle reinforcements in close relation to the matrix containing them The particles can be softer or harder compared with the matrix Ceramic particles generally used as reinforcements include carbon, silicon carbide, and alumina, which have low adhesion to a metallic counterface The asperity of the counterface can easily plow through softer particles such as char (carbon); it cannot do so through harder particles such as alumina or silicon carbide

Figure 5 shows the variation of specific wear rate as a function of volume fraction of particles in different systems (Ref

12, 32) Char is a soft particle that contains carbon and other hard mineral oxides, and composites containing char exhibit much wear Alumina is harder than char but softer than silicon carbide, and an aluminum-alumina composite shows higher wear than an Al-SiC composite but lower wear than an aluminum-char composite However, aluminum-base composites that contain graphite have the lowest wear, because graphite is not only soft but also shears easily along the basal plane of its hexagonal close-packed lattice in suitable environments and acts as a solid lubricant Composites containing a solid lubricant such as graphite have low wear because of its transfer onto the tribosurfaces and the subsequent formation of a lubricating film between the matrix of the composite and the counterface Thus, it may be concluded that the harder the reinforcing particles, the lower the sliding wear in a composite, with the exception of solid lubricating dispersoids of the right crystal structure, which impart low wear rate despite their low hardness Another interesting observation in Fig 5 is that wear resistance can be improved by increasing the volume fraction of either hard

or soft particles in the range of conditions investigated

Fig 5 Specific wear rates for several aluminum alloy particulate composites sliding against steel as a function

of particle volume fraction

The friction coefficient values for various composites are given in Fig 6 It seems that particles with higher hardness result in higher coefficients of friction Figure 6 can be divided roughly into three overlapping regions (Ref 24, 33, 35, 36,

37, 38, 39, 40, 41, 42, 43, 44):

hardness

However, this classification does not reflect matrix characteristics and thus can be applied only to composites of the same

or similar matrix The hard particles in composites result in larger plastic deformation due to asperity-asperity interaction during sliding and contribute to higher coefficients of friction despite reduced contribution of adhesion between the particle and the matrix In region 3, a lubricating film forms on the mating surface, caused by the smearing of soft solid lubricating particles during the sliding process In region 2, although a transfer layer may form during sliding, this layer cannot reduce the coefficient of friction significantly because it is not a lubricating film Solid lubricating films are responsible for lower friction coefficients in composites that contain solid lubricants such as graphite or MoS2 Rohatgi et

al (Ref 45) have published detailed results on the tribological behavior of metal-matrix/graphite-particle composites

Their results indicate that when the volume fraction of graphite in the composite is greater than 20 vol%, the coefficient

of friction is close to 0.2 regardless of the matrix Apparently, above 20 vol% graphite, the graphite film covers the entire surface and prevents direct metal-to-metal contact, as shown in Fig 7 (Ref 40, 42, 43, 45, 46, 47, 48, 49, 50) Under these conditions, the friction coefficient is not a significant function of the matrix alloy chemistry

Fig 6 Coefficient of sliding friction as a function of particle volume fraction in metal-matrix composites sliding

against a steel counterface (except those with a silver matrix) See text for discussion of regions 1 to 3

Normal pressure Symbol Material

MPa ksi

Sliding speed,

Fig 7 Coefficient of sliding friction as a function of volume fraction of graphite particles in metal-matrix

composites sliding against steel (except those with a silver matrix)

Friction and Abrasive Wear. The abrasion resistance of a material generally is determined by two different types of tests: (1) low-stress tests, such as rubber wheel abrasion tests (RWAT; for details, see ASTM Standards G-65 and B-611) and (2) high-stress tests, such as pin-on-disk and pin-on-drum types of tests In low-stress tests, the abrading particles do not break; in tests employing high stress levels, the abrading particles break The abrasive particles are generally rounded sand (SiO2), crushed quartz, or silicon carbide particles of different grit sizes and shapes The size and shape of abrading particles, as well as their relative hardness with respect to the composite being tested, are important parameters in determining the extent of wear Bhattacharrya and Bock (Ref 51) examined the effects of different abrasive particles on the abrasive wear of engineering materials and concluded that alumina is more aggressive than silica sand In monolithic materials, it has been observed (Ref 52) that wear volume increases rapidly with grit size of abrading particles up to a critical diameter; beyond this size, wear increases at a slower rate When the size of the abrading particle is less than 1

m, the wear is no longer by abrasion In metal-matrix composites, as in monolithic materials, abrasive wear involves gouging, grooving, and plastic deformation caused by penetration of hard abrading particles (Ref 53) The abrading particles may also interact with the dispersed hard particles in the composite In RWATs, the stress developed during this interaction between hard particles cannot exceed a limiting stress required for local elastic deformation of the rubber wheel to engulf the abrading particle In high-stress tests employing the rigid backing surface of a disk or drum, the particle interaction can be very severe, leading to widespread fracture of the hard but brittle particles

Specific abrasive wear rates as a function of volume fraction for several hard particles dispersed in aluminum matrices are shown in Fig 8 (Ref 14, 16, 37, 54) The wear rates decrease with an increase in volume fraction of particles in a manner

similar to those observed in adhesive wear For a given volume fraction of particles, composites that contain harder particles exhibit a lower wear rate Although the values of abrasive wear show considerable scatter, the specific wear rate for aluminum-base composites containing hard particles is 10-6 to 10-7 mm3/mm · N when the volume fraction of particles

is less than 40% Tawarada et al (Ref 55) have pointed out that the specific wear rate for Al-4.5Mg alloy containing more

than 50% Al2O3 particles is extremely low, of the order of 10-9 to 10-10 mm3/mm · N

Fig 8 Specific abrasive wear rate as a function of particle volume fraction in aluminum alloy composites

The improvement in abrasion wear rate of various metal-matrix composites over that of the base alloy is a function of the volume of different reinforcing particles when subjected to RWATs (Fig 9) (Ref 56) The properties of the dispersoid did not significantly affect these results, probably because of low-stress interaction between the dispersoid and the abrading particles at a level much lower than the hardness of the dispersoids Under high-stress conditions, the wear rate decreases almost linearly with the volume fraction of dispersoids in a composite Similar results have been reported for copper-base hard-particle composites (Ref 57) Apparently, the particle spacing decreases with an increase in volume fraction, and any cracks formed easily propagate through the reduced distance in the ductile matrix region to reach the neighboring particles; this results in faster wear of the composites (Ref 58, 59, 60)

Fig 9 Normalized abrasion rate as a function of particle volume fraction for aluminum alloy composites under

low-stress abrasive wear

Very few publications on steel-base particulate composites are available in open literature Recently, Halley et al (Ref

61) have reported that the abrasive wear volume of tool steel (0.6-0.9C-10Cr-Mo) with 17 vol% TiC particle composites was lower by 54% compared with similar tool steels Other investigators (Ref 62) also report that wear rate decreases with an increase of TiC volume fraction for steel-base TiC particle composites

Under abrasive wear, the coefficient of friction in composites containing hard particles is also a function of the volume fraction of dispersoids As the volume fraction of hard particles increases, the coefficient of friction decreases for both aluminum-base (Ref 42) and copper-base (Ref 57) composites The increase in the volume fraction of hard particles in the matrix alloy reduces the area fraction of matrix, and thus there is enhanced ceramic-ceramic contact The friction coefficient for ceramic-ceramic contact can be 0.25 or less (Ref 57, 63), and that for metal-metal contact is generally much higher (Ref 57, 64) The increase in the relative contact area involving ceramic particles increases the contribution

of ceramic-ceramic contacts, resulting in an overall decrease in friction coefficient

Friction and Erosive Wear. In most cases, the abrasive wear resistance of composites increases with increasing particle volume fraction of hard particles, as discussed above However, in the case of erosive wear (for details, see ASTM Standard G-76), opposite results have been reported (Ref 65, 66, 67, 68), as shown in Fig 10 The steady-state erosion rate for 75 to 200 mquartz erodent at 60 ms-1 and an impact angle of 60° increases with an increase of carbide content in both iron- and cobalt-base composites Similar results have been obtained for other impact angles In addition,

if the matrices are changed from metals to ceramics, the erosion rate decreases with the content of carbides in ceramics (Fig 10) Ball and Patterson (Ref 67) have pointed out that the minimum in erosive wear rate appeared at 34 vol% binder

in WC-Co cermets A similar result has been reported in the same system for slurry erosive wear (Ref 68) It has been established that, in erosive wear, the angle of impact is very important In ductile materials, erosive wear is maximum at

an impact angle of 15° to 30° for brittle materials, however, this angle is about 90° Goretta et al (Ref 69) have found that

erosive wear rates reach a maximum when the impact angle of the erodent Al2O3 was 15° to 30° in Al-20SiC particle composites, and the wear rates of the Al-SiC composites are higher than that of the base alloy under impact angles from 15° to 90° It is believed (Ref 65, 66, 67, 68, 69, 70) that erosion rates are affected by particle size, particle agglomeration, heat treatment, erodent hardness, impact angle and velocity, and volume fraction of hard particles (or second phase) It has been claimed that erosion rates of metallic alloys are not likely to be reduced by additions of hard particles, except for cases when relatively soft particles (such as quartz abrasive) are used as erodents (Ref 66)