Volume 07 - Powder Metal Technologies and Applications Part 9 ppsx

Huo, Determination of Threshold Pressure for Infiltration of Liquid Aluminum into Short Alumina Fiber Preform, Trans.. Huo, Determination of Threshold Pressure for Infiltration of Liqui

Trang 1

Fig 4 Schematic drawing of microstructure of tungsten sintered at different temperatures before being

immersed into molten copper (a) 1100 °C (2010 °F) (b) 1350 °C (2460 °F) (c) 1600 °C (2910 °F) (d) 2800

°C (5070 °F) 500× Source: Ref 16

Infiltrated bodies that have a continuous skeleton phase above approximately 65 vol% can be shaped only by machining, while those having a smaller proportion of refractory metal dispersed as loose grains in the ductile metal matrix are plastically deformable at elevated temperatures (Ref 8) Soldering, brazing, or plating of the infiltrated product is aided by generally smooth surface films or high contents of infiltrant metal

The preceding binary systems illustrate combinations of two metals of widely differing melting temperatures that can be advantageously produced to near-net shape and full density by infiltration Nickel also can be infiltrated into tungsten, but equilibrium at liquid-phase temperature causes severe attack of the refractory metal and requires careful process control to prevent incomplete penetration due to diffusion solidification Coarse tungsten powder helps to produce compacts with larger capillary channels for better penetration High heating rates, especially after about 90% of the absolute melting temperature of nickel is reached, improve infiltration conditions

Trang 2

If nickel is alloyed with copper, the solubility of tungsten in the liquid phase decreases as the copper percentage is increased Infiltration is more practical (Ref 8) for heavy alloy systems because tungsten powder and process control requirements are less stringent If nickel is alloyed with iron, however, the solid-liquid phase interaction is similar to the binary tungsten-nickel system, and infiltration again is more difficult

The same principle applies to the more complex refractory metal systems with nickel-chromium and cobalt-chromium alloys Nevertheless, skeleton bodies of tungsten and molybdenum, as well as bodies of binary 85W-15Cr and 75W-25Cr alloys, can be successfully infiltrated with superalloys of the Nichrome-V, Hastelloy-C, Stellite, and Vitallium compositions into shapes simulating mechanical and engine test specimens (Ref 17)

Several other refractory metal-based composite structures can be readily produced by infiltration These include the density tungsten-lead system to produce materials suitable for shielding against radiation and the chromium-copper system to produce compositions for welding electrodes (Ref 8) To retain the low liquid-solid contact angles in these systems, a strong reducing atmosphere is necessary to prevent oxide films on the molten lead or on the solid chromium A free metallic surface requires sintering of the skeleton above 1250 °C (2280 °F) to reduce any oxide film on the solid chromium

high-Carbide-Based Systems. Liquid-phase sintering of tungsten carbide/cobalt or titanium carbide/nickel systems capitalizes on the eutectics of the two phases A limited solubility of the carbide in the matrix metal facilitates bonding; carbon and metal diffusion through the liquid are less important in densification than reactions at the carbide-metal phase boundaries It is unknown whether the carbide particles in these systems form a rigid skeleton, but the interfacial tension between crystals of the carbide and the liquid metal appears to be anisotropic (Ref 18)

During cooling from the sintering temperature, some or most of the carbon and metal dissolved in the liquid precipitates

on grains that remained solid during the process This mechanism can be altered somewhat if the rigid skeleton is formed first and the liquid phase subsequently infiltrates into the pore system By first saturating the matrix metal with carbon and skeleton metal, the liquid phase dissolves less skeleton materials, and shape distortion is diminished Carbide coalescence and grain growth also are decreased

The earliest attempts to produce cemented carbides were made by infiltrating carbide skeletons with unalloyed binder metals (Ref 19) Later, binder metal prealloyed with elements of the skeleton to inhibit contact face erosion was used in a broad investigation of infiltrating single and double carbides with many cobalt and nickel alloys (Ref 20) Table 2 lists some of the alloys used The large number of feasible combinations includes several noteworthy successes, especially for titanium carbide-based stems with Nichrome and Vitallium infiltrants A feasibility study (Ref 17) produced similar results for the same infiltration systems and laid the foundation for an extensive development program to utilize these materials for heat-resistant applications

Trang 3

Table 2 Carbide infiltration test matrix and evaluation

%

Infiltrant composition,

%

Type

°C °F

Time, min

Very tough 1b

11.05- 14.6

29.3-60Co-40WC

Contact, opposing sides(d)

1350 2460 5

85-86 No erosion,

heavy residue

Porous in core, graphite precipitates, fairly uniform grain size

6.65- 21.3 95Co-5WC

37.1-Contact, opposing sides(d)

1460 2660 15

85.6-87 Contact face

erosion

Uniform phase distribution, some porosity

3b

72.7Co- 10TiC

3.01- 19.3

33.2-80Ni-20Cr

Contact, opposing sides(d)

1450 2640 15 24.6Ni,

6.1Cr, bal TiC

4b

72.7Co- 10TiC

17.3Cr-Contact, one side(d)

1400 2550 5

88+ Less contact

face residue than in 4a

Higher matrix concentration near contact face, porosity increasing toward far end

Tough where dense, brittle where porous

3.38- 11.4

25.7-80Ni-20Cr Capillary

dip in molten infiltrant

1550 2820 3 22.5Ni,

5.7Cr, 2.1Mo 2 C, bal TiC

84.5-85 Alloy skin

becoming heavier toward bottom, forming excess on bottom end

Uniform phase distribution, generally dense

3.46- 11.6

24.8-

72.7Co-Contact, one side(d)

Trang 4

10TiC

17.3Cr-contact face residue

72.7Ni- 10TiC

17.3Cr-Capillary dip in molten infiltrant

1550 2820 3 85 Similar to 4d Similar to 4d Tough 5d

90TiC-10Mo 2 C(e)

4.14

3.54- 14.6

26.0-80Ni-20Cr Contact,

opposing sides(d)

1400 2550 15 22.9Ni,

85-86 Slight

contact face erosion, small residue, slightly porous

Porous in core, less uniform phase distribution than in 4d

Less tough than 4 and 5

6

70TiC-30Mo 2 C(e)

4.75

4.09-22.9-9.9 80Ni-20Cr Contact,

opposing sides(d)

1400 2550 15 22.6Ni,

86-87 Similar to 6,

but more porous

More porous in core, less uniform phase distribution than in 6

More brittle than 6

7

50TiC-50Mo 2 C(e)

5.58

4.69-21.3-8.3 80Ni-20Cr Contact,

opposing sides(d)

1400 2550 15 22.3Ni,

5.7Cr, 35.8Mo 2 C bal TiC

86-87 Similar to 7,

but more porous

Very porous, nonuniform phase distribution

More brittle than 7

8

Source: Ref 19

(a) All skeletons were presintered at 950 °C (1740 °F) and high sintered at 1500 °C (2730 °F) for 2 h in a carbon tube resistor furnace under hydrogen,

except No 1 to 3, which were high sintered in vacua in a carbon susceptor induction furnace

(c) A qualitative assessment of resistance against fragmentation by hammer blows

(d) Infiltrant mass was 40 to 45% of mass of infiltrated product

Trang 5

The structure of infiltrated carbides reflects infiltration mechanics on a macroscale In zones penetrated by the infiltrant, fully dense regions and slight expansion of the skeleton due to carbide grain separation are observed Substantial porosity and some shrinkage occur in areas inadequately penetrated by the liquid alloy, such as the side opposite the contact face

in unidirectional infiltration, or in the center for infiltration from opposite sides Subsequent heat treatment is ineffective

in eliminating porosity Erosion at the contact faces is greatly diminished by infiltrating skeletons composed of base multicarbides, so that saturating the infiltrant with skeleton elements by prealloys frequently is not required

tungsten-The microstructure reflects the crystallographic characteristics of the carbide, and fully infiltrated regions do not differ in grain size and morphology from material whose liquid phase was sintered in situ Rectangular and triangular grains are retained in infiltrated tungsten carbide, whereas for infiltrated titanium carbide, the cubic lattice is reflected by distinctly rounded grains Grains of solid-solution carbides of tungsten and titanium or titanium and molybdenum are slightly rounded at the corners Graphite precipitates accompany porosity in poorly infiltrated regions for tungsten carbide and titanium carbide skeletons, especially if starting powders contain more than a trace of free carbon

Ferrous-Base Systems. The thermodynamic affinity between solid iron and liquid copper offers the potential for virtually pore-free P/M products by infiltration Moreover, the excellent wetting characteristics that exist in the brazing of steel can also be utilized for joining disparate bodies during infiltration A powder compact and a casting or forging, or halves of complex or offset configurations, can be joined Figure 5 (Ref 10) illustrates this self-brazing capacity without strength degradation Finally, the generally smooth cupric film surrounding the infiltrated body serves as a base for surface coating or plating

Fig 5 Infiltration-brazed butted iron-copper bars Green bars were clamped together end-to-end, and one free

end surface was contacted with molten infiltrant Rupture of composite bar occurred away from the joint, evidence of the high strength of the brazed bond (a) Butt joined bars after infiltration (b) Machined tensile bar (c) Tested tensile bar

The ability of the infiltration process to combine major proportions of normally unalloyable industrial premier metals (iron and copper) was recognized as early as World War I (Ref 21, 22), but it was only in the late 1940s, through refinements in technique, that sound products could be made (Ref 10, 23, 24, 25) These products, in turn, have culminated in the present advanced state of the art While copper content must be higher than for most commercially sintered iron-copper alloys because of the need to maintain an interconnected pore system for complete infiltration, good mechanical properties can be realized

This is apparent from the tensile strength-elongation data given in Fig 6 for commercial iron powder with and without graphite additions (Ref 24) Strength is enhanced because the infiltrated structure, with a minimal amount of isolated pores, is virtually free of internal notches The iron-copper system permits a precipitation-strengthening mechanism If carbon is diffused into the iron to produce a hypoeutectoid structure of the skeleton, hardening by martensite transformation is possible Where other metals are alloyed with the copper, solid solution strengthening of the matrix can

be achieved

Trang 6

Fig 6 Mechanical property ranges of copper-infiltrated iron and hypoeutectoid steel compacts before and after

heat treatment

In the binary iron-copper system, about 3.8 to 4% Fe is dissolved by the liquid copper under equilibrium conditions at an infiltration temperature of 25 to 50 °C (45 to 90 °F) above the peritectic temperature of 1090 °C (2000 °F) at the copper side of the phase diagram, while the -iron dissolves about 8 to 8.5% of the copper At 900 °C (1650 °F), the solubility

of -iron in copper is about 1.5%; it diminishes to less than 0.04% at room temperature, at which point the solubility of copper in iron is equally low

These thermodynamic relations form the basis for precipitation hardening However, macrodispersion of the two phases

in the infiltrated alloys causes concentration of precipitates in thin zones at the phase boundaries, as shown in Fig 7 (Ref 26) Consequently, conventional hardness tests are not precise enough to show a noticeable increase in macrohardness after a precipitation treatment, such as quenching from 900 °C (1650 °F) followed by prolonged tempering at 600 °C (1110 °F) The precipitation mechanism produces increases in strength, elongation, and impact resistance (Ref 8) and also can be traced through changes in the electrical conductivity (Ref 25)

Trang 7

Fig 7 Microstructure of 25Cu-Fe compact sintered at 1100 °C (2010 °F) for 30 min Large, rounded, dark areas

are -iron, separated by a diffusion layer from the light copper phase containing the fine, dark, iron-rich precipitate 1000× Source: Ref 26

If carbon is diffused into the iron skeleton with copper infiltrated afterward, two processes compete with one another during cooling During quenching, normal martensitic transformation occurs inside the skeleton structure At the same time, however, precipitation of the dissolved iron in the copper and copper in iron, respectively, is suppressed in the phase boundary zones During reheating, martensite decomposition causes a decrease in hardness of the steel skeleton structure, with simultaneous increases in hardness and strength of the boundary zones

Table 3 shows the effect of such a heat treatment on the mechanical properties of a copper-infiltrated 0.3% carbon steel Water quenching produces an appreciable increase in hardness and brittleness With increasing reheating temperatures, however, the material becomes softer and tougher, without loss in strength (Ref 8)

Table 3 Effect of heat treatment on the mechanical properties of 0.3% carbon steel infiltrated with 11 vol% Cu

Ultimate

tensile strength

Impact resistance (b)

Reheated to 900 °C (1650 °F) and water quenched 437 790 114.6 5.2 11.76 8.68

Reheated for 2 h at 400 °C (750 °F) 360 14.71 10.85

Reheated for 2 h at 500 °C (930 °F) 302 771 111.8 7.6 23.14 17.07

Reheated for 2 h at 600 °C (1110 °F) 255 750 108.8 13.5 62.75 46.30

(a) Hametag iron powder pressed to 88% of theoretical density, sintered at 1220 °C (2230 °F) for 1 h in

hydrogen, and infiltrated with electrolytic copper at 1100 °C (2010 °F) for 30 min

(b) Unnotched test bar of 1 cm2 cross section

When a copper alloy is used as an infiltrant, other benefits can accrue When precipitates form in the matrix, such as copper alloys with beryllium, chromium, or silicon, the resulting strength increase during heat treating of the infiltrated body reinforces iron-copper precipitation zones at the boundaries The strengthening effect of precipitates is independent

of heat treatment and augments strength increases obtained with unalloyed copper infiltrant in solid-solution alloys However, some solid-solution alloy infiltrants minimize erosion at the point of initial contact with the iron skeleton, because these alloys melt slowly over a range of temperatures Brass containing 20% Zn exhibits this phenomenon

Trang 8

A copper alloy with manganese in amounts up to 5%, especially if it also contains sufficient iron to inhibit a severe attack

of the skeleton contact faces, has a different beneficial effect The manganese oxidizes preferentially in the commercial atmospheres where infiltration takes place A nonadhering porous crust results that can be removed during finishing more easily than the tenacious residue formed by a binary copper-iron infiltrant (Ref 27, 28)

Many ferrous metal-base infiltration systems have been explored experimentally Kieffer and Benesovsky (Ref 8) have investigated the iron-gold, iron-bismuth, iron-cadmium, iron-lead, iron-antimony, and iron-tin systems for bearings, and the iron-cobalt-silicon, iron-copper-silicon, and iron-manganese-silicon systems for magnetic or structural parts Alloys of the iron-zinc system have also been produced by infiltration, but treatment in a pressure vessel is required to overcome the high vapor pressure of the zinc (Ref 29)

Austenitic stainless steel skeletons infiltrated with silver possess excellent corrosion resistance, thus making them suitable for food processing applications Ferritic stainless steel and high-manganese steel compacts with varying carbon contents also display improved corrosion resistance when infiltrated with cupric alloys These alloys also exhibit extraordinary hardness and wear resistance, coupled with a considerable toughness Table 4 lists mechanical and technological properties of several ferrous-base infiltrated materials

Trang 9

Table 4 Properties of some ferrous metal-based infiltrated alloys

tensile strength

Impact resistance(a)

Skeleton

composition,

%

Infiltrant composition,

93.2Fe-6Mn-0.8C 100Cu 13 7.90 7.87 740 Naturally hard, wear resistant

87.2Fe-12Mn-0.8C 100Cu 9 7.89 7.69 310 562 81.5 6 Wear resistant, work hardening

93.5Fe-3Cr-3Mn-0.5C 100Cu 14 7.96 7.93 502 957 138.8 4

Trang 10

Nonferrous-Based Systems. The major nonferrous metals with higher melting points are thermodynamically compatible with many low-melting metals in the liquid state Consequently, skeleton bodies of cobalt and nickel can be readily infiltrated with gold, as well as with many of the low-melting heavy metals, such as bismuth, lead, or antimony (Ref 8) Copper can also be infiltrated into cobalt and nickel skeletons; however, because of the formation of solid solution in all proportions, infiltration into nickel powder compacts requires a narrow particle size range, wide capillaries within a pore volume not exceeding about 35%, short infiltration time, and a vacuum to assist the capillary forces Mercury wets nickel well without forming an amalgam and is easily impregnated into skeleton bodies, provided the pore structure prevents exudation of the heavy liquid metal (Ref 30)

Copper is another skeleton metal with pores that can be readily filled with liquid low-melting-point metals, such as lead (Ref 31, 32) or bismuth (Ref 33) Vacuum impregnation is suitable for incorporating lead-base alloys, such as those containing 15% Sb and 5 to 10% Sn, into spongy structures of nickel-copper or nickel-iron supported by steel backing (Ref 34)

Combinations of aluminum or aluminum alloys with low-melting metals such as bismuth, lead, thallium, or thallium-lead

by means of infiltration in hydrogen or a vacuum have been proposed (Ref 33, 35) However, strict control of powder characteristics, especially particle shape and surface condition, to maximize wetting appears to put infiltration at a disadvantage over in situ liquid-phase sintering To overcome this problem, zinc or cadmium can be added to the aluminum of the skeleton, followed by cleaning and activating the free surface of the pores by evaporation of the lower boiling metal before proceeding with the impregnation of a metal such as lead (Ref 36)

References cited in this section

8 R Kieffer and F Benesovsky, The Production and Properties of Novel Sintered Alloys (Infiltrated Alloys),

Berg- und Hüttenmännische Monatshefte, Vol 94 (No 8/9), 1949, p 284-294

10 F.V Lend, Powder Metallurgy, Metal Powder Industries Federation, 1980, p 313-319

15 C.G Goetzel, Treatise on Powder Metallurgy, Vol 2, Interscience, 1950, p 196

16 K Schröter, Border Regions of Metallography, Z Metallkd., Vol 23 (No 7), 1931, p 197-201

17 J.M Krol and C.G Goetzel, "Refractory Metal Reinforced Super Alloys," USAF Technical Report 5892, ATI No 57154, May 1949

19 R Kieffer and F Kölbl, Production of Hard Metals by Infiltration, Berg Hüttenmämann Monatsh., Vol 95

22 C.L Gebauer, Process of Producing Metal Bodies, U.S Patent 1,342,801, 1920; C.L Gebauer, Production

of a Composite Metallic Article, U.S Patent 1,395,269, 1921

23 F.P Peters, Cemented Steels A New High-Strength Powder Metallurgy Product, Materials and Methods,

Vol 23 (No 4), 1946, p 987-991

24 E.S Kopecki, Cemented Steels, Iron Age, Vol 157 (No 18), 1946, p 50-54

25 C.G Goetzel, Cemented Steels Infiltration Studies with Pure Iron and Copper Powders, Powder Metall Bull., Vol 1 (No 3), 1946, p 37-43

26 L Northcott and C.J Leadbeater, Sintered Iron-Copper Compacts, Symposium on Powder Metallurgy,

Special Report No 38, The Iron and Steel Institute, 1947, p 142-150

27 P Schwarzkopf, Infiltration of Powder Metal Compacts with Liquid Metal, Met Prog., Vol 57 (No 1),

Trang 11

Vol 39, 1948, p 71-78

30 F.R Hensel, Treatment of Bearing, U.S Patent 2,364,713, 1941

31 E Fetz, Bearings from Metal Powder A New Art, Metals and Alloys, Vol 8 (No 9), 1937, p 257-260

32 E Fetz, Manufacture of Composite Bearings, U.S Patent 2,234,371, 1941

33 F.R Hensel, Impregnation of Metallic Composition with Bismuth, British Patent 590,412, 1947

34 A.L Boegehold, Copper-Nickel-Lead Bearings, Powder Metall., J Wulff, Ed., American Society for

Metals, 1942, p 520-529

35 F.R Hensel and E.I Larson, Sintered Porous Aluminum-Base Bearings, U.S Patent 2,418,841, 1947

36 F.R Hensel, Method of Making Porous Bearing Surfaces, U.S Patent 2,447,980, 1948

Because the two metal types do not alloy with one another, the rule of mixture can be applied to determine the density of

a specific composition Figure 8 shows the effect of increasing tungsten content on density, hardness, and electrical conductivity for copper-tungsten contact material (Ref 37) The straight-line relationship between volume ratio and conductivity is typical for this system Figure 9 shows the change in thermal expansion coefficient with tungsten content; principal mechanical properties versus tungsten content are plotted in Fig 10 (Ref 38) The boldface portions of the curves represent composition ranges of material that can be readily produced by infiltration with a low-melting metal content of approximately 10 to 40 vol% (5 to 25 wt%)

Trang 12

Mixing of the component elemental powders, briquetting, sintering, subsequent working

Very fine, 1-50 •

Impregnation of the loosely filled tungsten powder in a mold, extrusion into shape Coarse, 100-400

Coarse, 100-400 Medium, 50-100

Dip impregnation of a pressed (and presintered) tungsten compact, machining to size

Very fine, 1-50

Fig 8 Effect of composition on physical properties of tungsten-copper contact material (a) Density (b)

Hardness (c) Electrical conductivity Heavy framed areas are composition ranges for liquid copper infiltration or sintering of compacts from mixtures Source: Ref 37

Trang 13

Fig 9 Effect of composition on thermal expansion of tungsten-copper contact material Boldface portion of

curves designates liquid copper-infiltrated tungsten or sintered compacts from powder mixtures Light portion

of curves designates powder mixtures; compositions are not infiltrable Source: Ref 38

Fig 10 Effect of composition on mechanical properties of tungsten-copper contact material Boldface portion of

curves designates liquid copper-infiltrated tungsten or sintered compacts from powder mixtures Light portion

of curves designates powder mixtures; compositions are not infiltrable HRB Source: Ref 39

Trang 14

Table 6 lists mechanical and electrical property data for tungsten-silver and molybdenum-silver contact materials (Ref 39) Also included are data for nickel-silver contact materials that can be produced by infiltration

Table 6 Properties of contact materials containing silver

Composition, wt% Transverse rupture strength Electrical conductivity

Ag W Mo Ni

Density, g/cm3

Hardness,

HB MPa ksi Mmho/cm %IACS

Estimated contact resistance(a)

(b) 30T68 Hardness, Rockwell Superficial

(c) 30T46 Hardness, Rockwell Superficial

In infiltrated materials, the soft conductor metal matrix is the minor constituent, and extrusion for profile generation is not practical except for nickel-silver composites All materials are readily machinable into a variety of shapes, especially caps, platelets, and rings Usually, the heavy-duty contact metal, as a facing, is joined by clamping or brazing onto a support structure The facing also may be penetrated by an infiltrant with a major portion, that on cooling, solidifies like a casting; an example is copper that is strengthened by small amounts of beryllium, chromium, or nickel and is cast into the form of a backing, arm, or other support structure

Primary applications for infiltrated tungsten-copper composites are resistance welding electrodes and make-and-break contact facings in oil or air circuit breakers or transformer taps Tungsten-silver contacts are used in switch gear and low-voltage regulators (Ref 37) Low contact resistance and nonsticking properties of nickel-silver contacts make them suitable for high-voltage disconnect switches (Ref 39) Porous, cup-shaped nickel compacts have been impregnated with mercury as part of a short circuiting safety switch contact in proximity fuses for radio-controlled detonation (Ref 40, 41)

Rocket Nozzles. The manufacture of silver-infiltrated tungsten contacts formed the basis for the development and production of large billets, from which rocket nozzle throat liners have been fabricated This material has proved successful in actual service in rockets of underwater-launched ballistic missiles (Ref 42) because:

• Extreme stability of the structure (virtually zero change in critical dimensions due to the temperature, high-pressure gaseous environment in the nozzle throat)

high-• High resistance to thermal shock during heating and cooling at the extremely high rates experienced by the throat surface in contact with rocket propulsion gas This in turn requires high tensile strength and resistance to high hoop stresses at the outer perimeter of the throat liner, as well a good thermal conductivity to provide adequate heat transfer to the backup structure

• Fabricability, especially machinability, on an economical production scale

• Optimum reliability and reproducibility of the nozzle throats in spite of their relatively large size, compared to similarly produced P/M parts in electrical switch gear

An additional advantage of using silver as the matrix metal is its relatively high vapor pressure Silver evaporation aids infiltration of large capillaries by forming deposits on the pore walls before the liquid metal enters individual pores Also,

Trang 15

the exposed surface region of the throat liner is kept at a temperature considerably below that of the propulsion gas due to evaporative transpiration cooling (Ref 43)

Figure 11 shows a typical silver-infiltrated tungsten ring machined from a large billet that weighed nearly 90 kg (200 lb)

It contained about 80 vol% W; the conductor metal was uniformly dispersed throughout the entire cross section, filling all interconnected pores (Fig 12) Processing parameters, production details, and starting material characteristics are described in Ref 44 and 45

Fig 11 Silver-infiltrated tungsten billet for rocket nozzle throat liner Source: Ref 42

Fig 12 Microstructure of silver-infiltrated 80W-20Ag billet Angular microconstituents are silver 500× Source:

Ref 42

Toensing and Zalsman (Ref 44) have also reported on the effect of these parameters on mechanical properties at ambient and elevated temperatures Figure 13 shows the change in tensile strength with temperature for a tungsten skeleton with 20% pore volume before and after silver infiltration Data represent average values, but strengths as high as 620 to 700 MPa (90 to 100 ksi) have been reported As the melting temperature of the silver is approached, the strength of the composite converges with that of the tungsten skeleton Table 7 summarizes mechanical and physical properties of 20 vol% silver-infiltrated material for different test temperatures within the operating temperature range experienced by the throat liner during rocket propulsion (Ref 42)

Trang 16

Table 7 Physical and mechanical properties of silver-infiltrated tungsten

Determined at strain rate of 50 m/mm/min (0.05 in./in./min) for 20 vol% Ag/80 vol% W+

strength

Ultimate tensile

strength

Ultimate shear

Fig 13 Effect of temperature on tensile strength of porous uninfiltrated and silver-infiltrated (20 vol%)

tungsten Source: Ref 44

Jet Engine Components. The quest for materials to withstand the high temperatures and stresses imposed by combustion gases in the gas turbines of jet engines led to the development of cermets in the late 1940s and early 1950s Cermets offered increased operating temperatures and engine efficiency when used in rotating blades Their inherent brittleness prevented the use of cermets for this purpose, however, and their use for stationary gas-conducting nozzle vanes also failed to win acceptance For more information, see the article "Cermets and Cemented Carbides" in this Volume

To overcome brittleness, the infiltration process was applied to the production of gas turbine components Sintered and preformed titanium carbide skeletons were infiltrated with a nickel-or cobalt-base superalloy in a vacuum Graded products that had the following structural characteristics were produced:

Trang 17

• Ductile and tough roots and airfoil tips containing nearly 100% superalloy that withstood combined tensile and bend stresses at moderately high temperatures

• Ductile and tough superalloy-rich leading and trailing edges that resisted impact from small solid particles at temperatures ranging from ambient to operational

• Strong, creep-resistant airfoil portions containing 60 to 80 vol% titanium carbide that withstood the centrifugal force-induced tensile stresses at the highest temperature zones near the radial centroid of the foil, about midway between the root and tip

• Oxidation-protective, ductile, superalloy-rich airfoil encasements integrally joined with the metallic matrix

Figure 14(a) shows a graded turbine bucket for a J-47 jet engine after infiltration Figure 14(b) shows the graded turbine bucket after machining of the root configuration The change in microstructure across the airfoil is shown in Fig 15 (Ref 43) Table 8 gives stress-rupture for different zones of the graded bucket (Ref 43, 49) Other infiltrated titanium carbide turbine components are shown in Fig 16

Table 8 Stress-rupture properties of infiltrated graded cermet bucket

Airfoil tip Superalloy-rich region 870-930 1600-1700 55 8 870 1600 138 20

Airfoil body Superalloy-infiltrated TiC 1000 1800 70-100 10-15 980 1800 83 12

Fig 14 Graded cermet turbine blade (a) After superalloy infiltration of titanium carbide skeleton (b) After

machining of root

Trang 18

Fig 15 Microstructure of graded turbine blade across airfoil Angular and rounded microconstituents are

titanium carbide 150× Source: Ref 43

Fig 16 Infiltrated titanium carbide cermet jet engine turbine components (a) J-35 nozzle vane (b)

Experimental hollow vane (c) J-57 turbine bucket (d) J-47 turbine bucket

While these infiltrated carbide turbine components, like their sintered cermet counterparts, did not reach commercial production, blades for several stages of the compressor for a J-33 jet engine were mass produced from copper alloy infiltrated steel compacts (Ref 50) These blades were heat treated to a yield strength of 620 MPa (90 ksi), with an

Trang 19

elongation of 5%, and withstood three to four times as much vibration at an operating temperature of 370 °C (700 °F) as blades for these stages made from martensitic type 403 stainless steel precision forgings

Tools. In parallel development with turbine blades, titanium carbide skeletons were infiltrated with a variety of liquid steel alloys to manufacture tools and wear-resistant parts The infiltrant varied from simple low-carbon steel to alloy and high-speed steels (Ref 51) Angular or rounded titanium carbide grains resulted, depending on carbide content and matrix composition Infiltrated materials can be heat treated to produce specific properties For instance, hardness ranged from 90.1 to 90.6 HRA after water quenching and from 86.0 to 90.3 HRA after tempering At temperatures up to 750 °C (1380

°F), hot hardness for the titanium carbide that was infiltrated with type T6 tungsten high-speed steel was equivalent to commercial grades of cemented carbides This material had better oxidation resistance than cemented carbides up to 870

°C (1600 °F) and room-temperature transverse-rupture strengths up to 1500 MPa (220 ksi) Cutting speeds were twice those of molybdenum high-speed steel, one and one-half times those of Stellite, but only one fourth those of commercial steel-cutting grades of cemented carbides

Because the solubility of titanium carbide in liquid steel is high, the rate of infiltration is slowed so that penetration of the liquid tends to be confined to the regions near the original contact face The remaining porosity cannot be filled because

of diffusion solidification, regardless of the time allowed for penetration of the liquid As a result, severe size limitations are required for the infiltrated product to be sound and uniform in structure This disadvantage has caused the substitution

of in situ liquid-phase sintering of titanium carbide and steel powder mixtures in the production of such tool materials For more information, see the article "Cermets and Cemented Carbides" in this Volume

Mechanical Parts. Infiltration is widely used in the production of ferrous structural parts requiring densities in excess

of 7.4 g/cm3 and mechanical properties superior to those obtained by compacting, sintering, and coining Depending on the application, porous skeletons of iron or steel can be fully or partially infiltrated with copper alloy

There are several advantages to infiltration of iron-base structural parts with copper alloys (Ref 52):

• Increased mechanical properties Higher tensile strengths and hardnesses, greater impact energies, and

fatigue strengths are obtained through infiltration Figure 17 shows the effect of infiltration on strength

• Uniform density Parts that contain non-uniform and/or heavy sections can be infiltrated to obtain more

uniform density; infiltration tends to even out density variations

• Higher density Infiltration is a useful method to increase sintered part weight without increasing the

size of the part Given the press size limitations and restrictions in pressing technique and powder compressibility, it is often easier to obtain high density through infiltration Certainly, when considering normal P/M operations, it would be difficult to obtain densities in excess of 7.2 g/cm3 without resorting

to additional pressing and sintering operations Infiltration makes densities in excess of 7.2 g/cm3possible in a single pressing and sintering operation

• Removal of porosity for secondary operations Infiltration can be used in place of impregnation as a

method to seal surface porosity so that secondary operations such as pickling and plating can be performed without damaging the interior of the part and creating subsequent "bleeding" problems It is also a method of sealing a part used for applications in which no porosity is desired

• Selective property variation It is possible by infiltrating only selected areas of a part to obtain, within

limits, a controlled variation of properties in the part for example, variations in density, strength, and hardness This is known as localized infiltration Infiltration to considerably less than the full density in the part (e.g., 7.1 g/cm3) is known as starve infiltration

• Assembly of multiple parts Different sections of the final part, pressed separately, can be assembled by

sintering the individual pieces together and bonding the pieces into one part through common infiltration

Trang 20

Fig 17 Effect of infiltration on transverse-rupture strength of iron-carbon alloys sintered to a density of 6.4

g/cm 3 Combined carbon in alloys was about 80% of graphite added to iron powder; amount of copper infiltrant was adjusted to fill various fractions of void space Source: Ref 53

Table 9 gives composition and typical properties of P/M infiltrated steels

Table 9 Composition and properties of P/M infiltrated steels

MPIF composition limits(b), %

80.5-830 120 740 107 1.0 9.5 7.0 35 HRC 135 20

AS 0.6-1.0

8.0-14.9

91.4

80.1-895 130 725 105 60.5 9.5 7.0 40 HRC 135 20

FX-2000 (d) AS 0.3 max

15.0-25.0

85.0

70.7-450 65 1.0 20 15 60 HRB

AS 0.3-0.6

15.0-25.0

84.7

70.4-790 115 655 95 <0.5 8.1 6.0 30 HRC 125 18

AS 0.6-1.0

15.0-25.0

84.4

70.0-860 125 740 107 <0.5 6.8 5.0 42 HRC 125 18

Trang 21

Note: All materials have a density range of 7.2 to 7.6 g/cm

(a) AS, as sintered; HT, heat treated (typically austenitized at 870 °C (1600 °F)), oil quenched and tempered 1

h at 200 °C (390 °F)

(b) MPIF Standards require that the total amount of all other elements be <2.0%

The usual method of infiltrating iron and steel skeletons is to place a compact pressed from the powder of the infiltrant material next to the skeleton The compact of infiltrant powder can be positioned on top or underneath the skeleton compact, or two infiltrant compacts can be used one on top, the other underneath the skeleton compact The exact amount of infiltrant needed can be compacted in the same die in which the powder for the porous skeleton is pressed

After the skeleton has been sintered, the green compact or compacts of infiltrant powder are positioned next to the skeleton and the assembly heated to the infiltration temperature Sintering of the skeleton and infiltration can be combined into one operation, in which the green compact or compacts of the infiltrant are positioned next to the green compact of the skeleton By controlling the rate of heating, the skeleton compact will be adequately sintered by the time the melting point of the infiltrant is reached This operation is called "sintrating."

If only part of the porous skeleton is to be infiltrated for example, the teeth of an infiltrated gear the skeleton can be positioned in a graphite container in which space is provided adjacent to the gear teeth to be preferentially infiltrated This space is then filled with the appropriate amount of infiltrant in powder form Because iron has some solubility in copper, the liquid infiltrant will attack the surface of the skeleton when it first comes in contact with it, and severe erosion can take place at this point One method to minimize erosion is to use a copper alloy (e.g., 80%Cu-20%Zn brass) as infiltrant Because the brass does not melt at one temperature, but over a range of temperatures, less erosion takes place

Another method is to use an alloy of copper with iron as an infiltrant However, it is difficult to specify the exact amount

of iron; too little iron can cause erosion, and too much iron can cause an undesirable adherent deposit on the infiltrated part A third method is to use an alloy of copper, iron, and a third alloying constituent, such as manganese, which oxidizes

in the atmosphere in which infiltration takes place In this case, a crust containing an oxide of the oxidizing alloying ingredient of the infiltrant is formed, which does not adhere to the skeleton but which can be more or less readily removed

Figure 18 shows a small foot holder that required strength, machinability for tapping a thread, and plateability These characteristics were achieved by infiltrating a sintered iron-rich skeleton (7% Cu) with a copper-rich alloy containing 5%

Mn and 5% Fe The compacts were infiltrated under furnace conditions similar to those used for sintering, using infiltrating slugs previously pressed The infiltrated parts were electroplated with chromium to a satin finish Processing details for the two-step procedure are given in the table that accompanies Fig 18

Trang 22

Press and tool details

Type of press Mechanical(a)

Press capacity 45.4 metric tons (50 tons)

Die material Tungsten carbide (6% Co)

Punch material D2 tool steel

Core-rod material Tungsten carbide (6% Co)

Processing details

Compacting pressure: 415 MPa (60 ksi)

Infiltrating slugs 276 MPa (40 ksi)

Preheating treatment 15 min at 1100 °C (2040 °F)

Sintering treatment 15 min at 1100 °C (2040 °F)

Infiltrating treatment 15 min at 1100 °C (2040 °F)

Atmosphere for sintering and infitrating Endothermic

(a) Two movements above and three below

Fig 18 Small compact infiltrated with copper alloy to provide machinability, strength, and plateability

The control of dimensions during infiltration can cause problems It depends on the composition of the steel; carbon-free iron shows the greatest growth during infiltration It also depends on the exact temperature of infiltration and the time during which the infiltrated compact is above the liquidus temperature of the infiltrant

Bearings. The best example for the industrial use of infiltration in the bearing field is strip-backed steel, precision-type main, and connecting rod bearings for automobiles In this application, the anti-friction babbitt alloy is used to infiltrate the pores of a skeletal cupronickel strip that is bonded to the steel backing and produces an extra layer on the surface facing the crankpin (Ref 7, 34)

Infiltration techniques can also be applied to high-temperature bearing materials, such as bearing retainers in gas turbines for supersonic aircraft To combat the high wear of these cages, an ideal structure consists of a hard, load-bearing phase

Trang 23

intertwined with a liquid metallic lubricant Structures that closely approach the ideal concept and operate up to 370 °C (700 °F) have been developed (Ref 53)

Materials exhibiting improved anti-friction properties against steel under simulated service conditions involving very high rotational speeds are efficiently fabricated by P/M techniques that include an infiltration step These materials are characterized by a duplex structure (a soft, metallic phase of about 15 vol%) that is uniformly dispersed throughout the hard matrix The matrix can be either Monel or a 48Ni-48Cr-4Si alloy, with 10 wt% molybdenum disilicide added in some cases to reduce wear The soft phase consists of silver that is infiltrated after sintering the hard alloy powder compact

7 T Kimura, J.C Kosco, and A.J Shaler, Detergency During Infiltration in Powder Metallurgy, Proc of 15th Annual Meeting, Metal Powder Industries Federation, 1959, p 56-66

Metals, 1942, p 520-529

37 R Kieffer and W Hotop, Powder Metallurgy and Sintered Materials, Springer-Verlag, 1943, p 324-326,

329

38 F.R Hensel, E.I Larsen, and E.F Swazy, Physical Properties of Metal Compositions with a Refractory

Metal Base, Powder Metall., J Wulff, Ed., American Society for Metals, 1942, p 483-492

39 C.G Goetzel, Treatise on Powder Metallurgy, Vol 2, Interscience, 1950, p 207-209, 216-217

40 P Schwarzkopf, Powder Metallurgy, Macmillan, 1947, p 167-168

41 C.G Goetzel, Treatise on Powder Metallurgy, Vol 2, Interscience, 1950, p 539-540, 634

42 C.G Goetzel and J.B Rittenhouse, The Influence of Processing Conditions on the Properties of

Silver-Infiltrated Tungsten, Symposium sur la Métallurgie des Poudres, Edictions Métaux (Paris), 1964, p 279-288

43 C.G Goetzel and H.W Lavendel, Infiltrated Powder Components for Power Plant and Propulsion Systems,

Metals for the Space Age, Plansee Proceedings 1964, F Benesovsky, Ed., Springer-Verlag, 1965, p

46 J.R Kattus, Report, 5525-1428 II, Southern Research Institute, 19 Sept 1962

47 J.R Kattus, Report, 6253-1526 VI, Southern Research Institute, 9 Dec 1963

48 J.R Kattus, Report, 6606-1498 XI, Southern Research Institute, 31 Jan 1964

49 H.W Lavendel and C.G Goetzel, "A Study of Graded Cermet Components for High Temperature Turbine Applications," Report, WADC-TR 57-135, May 1957

50 G Stern and J.A Gerzina, Making Jet Engine Compressor Blades by Powder Metallurgy, Iron Age, Vol

165 (No 8), 1950, p 74-77

51 C.G Goetzel and L.P Skolnick, Some Properties of a Recently Developed Hard Metal Produced by

Infiltration, Sintered High-Temperature and Corrosion-Resistant Materials Plansee Proc., F Benesovsky,

Ed., Pergamon Press, 1956, p 92-98

52 C Durdaller, "Copper Infiltration of Iron-Based P/M Parts," Hoeganaes Corp., Riverton, NJ, 1969

53 J.T Burwell, Wear Behavior of High Temperature Bearing Materials, Prec Met Mold., Vol 14 (No 10),

1956, p 40, 41, 87, 88, 90, 91

Trang 24

on the latter topics is given in "Selected References" at the end of this article

Mechanism of Infiltration

Analytical treatments to describe fluid flow and heat transfer during infiltration of porous preforms by a liquid metal have been addressed by different groups at MIT (Ref 54, 55, 56, 57, 58, 59, 60, 61) and Colorado School of Mines (Ref 62) in the United States, also in England (Ref 63), France (Ref 64), Japan (Ref 65), China (Ref 66), and Jordan (Ref 67) General expressions have been derived to describe heat, mass, fluid flow, and kinetics during the infiltration process A review of theoretical factors and mechanisms dealing with infiltration of a porous body by a liquid metal is given in Ref

68 Factors that control the matrix grain size, interfacial reactions, and morphology of the fiber-matrix interface and its stability have been predicted (e.g., initial fiber temperature, preform compressibility, reaction heat, external cooling, or reaction kinetics) (Ref 54, 57, 59, 70) Infiltration kinetics, size of the remelting region, and temperature distributions can

be calculated (Ref 54, 58, 61, 62, 65, 70)

Models for pressure effects in conventional or new infiltration processes (squeeze casting or chemical vapor infiltration) have been developed (Ref 54, 57, 60, 61, 67, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80) Based on the analysis of physical phenomena that govern the infiltration process, good practical guidelines to optimize the processing techniques and the materials produced can be found in Ref 75

The mechanism of wetting and spreading of the liquid metal has been studied (Ref 54, 78, 81, 82) These studies reveal the formation of a precursor film ahead of the bulk liquid front by adsorption, condensation, and surface migration on the solid wall (Ref 82) A model for chemical vapor infiltration is presented in Ref 83 and 84

Sensors and models apparatuses (e.g., noninvasive capacitance technique), model experiments, and instrumented casting facilities have been developed to confirm theoretical predictions (Ref 55, 56, 58, 60, 64, 78, 81, 85, 86, 87) Modeling of magnetic field assisted and ultrasonic infiltration also has been attempted (Ref 88, 89, 90) A validation of the Lorenz force model has been experimentally obtained (Ref 88)

54 A Mortensen, L.J Masur, J.A Cornie, and M.C Flemings, Infiltration of Fibrous Preforms by a Pure

Metal, Part I: Theory, Metall Trans A, Vol 20, 1989, p 2535-2547

55 J Masur, A Mortensen, J.A Cornie, and M.C Flemings, Infiltration of Fibrous Preforms by a Pure Metal,

Part II: Experiment, Metall Trans A, Vol 20, 1989, p 2549-2557

56 A Mortensen, Infiltration of Fibrous Preforms by a Pure Metal, Part III: Capillary Phenomena, Metall Trans A, Vol 21, 1990, p 2257-2263

57 A Mortensen and V.J Michaud, Infiltration of Fiber Preforms by a Binary Alloy, Part I: Theory, Metall Trans A, Vol 21, 1990, p 2059-2072

58 V.J Michaud and A Mortensen, Infiltration of Fiber Preforms by a Binary Alloy, Part II: Further Theory

and Experiments, Metall Trans A, Vol 23, 1992, p 2263-2280

59 R.B Calhoun and A Mortensen, Infiltration of Fibrous Preforms by a Pure Metal, Part IV: Morphological

Trang 25

Stability of the Remelting Front, Metall Trans A, Vol 23, 1992, p 2291-2299

60 V.J Michaud, L.M Compton, and A Mortensen, Capillarity in Isothermal Infiltration of Alumina Fiber

Preforms with Aluminum, Metall Trans A, Vol 25, 1994, p 2145-2152

61 A Mortensen and J.A Cornie, On the Infiltration of Metal Matrix Composites, Metall Trans A, Vol 18,

1987, p 1160-1163

62 G.P Martins, D.L Olson, and G.R Edwards, Modeling of Infiltration Kinetics for Liquid Metal Processing

of Composites, Metall Trans B, Vol 19, 1988, p 95-101

63 R.M.K Young, Liquid Metal Infiltration Model of Unidirectional Fiber Preform in Inert Atmospheres,

Mater Sci Eng A, Vol 135, 1991, p 19-22

64 E Lacoste, M Aboulfatah, M Danis, and F Girot, Numerical Simulation of the Infiltration of Fibrous

Preforms by a Pure Metal, Metall Trans A, Vol 24, 1993, p 2667-2678

65 M Yokota, Basis and Application of Infiltration Techniques for Powder Metallurgy, J Jpn Soc Powder Powder Metall., Vol 38, 1991, p 464-471

66 L Hu, S Luo, and W Huo, Determination of Threshold Pressure for Infiltration of Liquid Aluminum into

Short Alumina Fiber Preform, Trans Nonferrous Met Soc China, Vol 6, 1996, p 133-137

67 A.E.M Assarand and M.A Alnimr, Fabrication of Metal Matrix Composites by Infiltration Process, Part I:

Modeling of Thermodynamic and Thermal Behavior, J Compos Mater., Vol 28, 1994, p 1480-1490

68 R Lumpkins, Theoretical Review of the Copper Infiltration of P/M Components, Powder Metall Int., Vol

17, 1985, p 120-123

70 X Tong and J.A Kahn, Infiltration and Solidification/Remelting of a Pure Metal in a Two-Dimensional

Porous Preform, J Heat Transfer (Trans AIME), Vol 118, 1996, p 173-180

71 T.W Cline and J.F Mason, Squeeze Infiltration Process of Metal-Matrix Composites, Metall Trans A, Vol

18, 1987, p 1519-1530

72 S Nourbakhsh, F.-L Liang, and H Margolin, Calculation of Minimum Pressure for Liquid Metal

Infiltration of a Fiber Array, Metall Trans A, Vol 20, 1985, p 1861-1866

73 N.H Tai and T.W Chou, Modeling of an Improved Chemical Vapor Infiltration Process for Ceramic

Composites Fabrication, J Am Ceram Soc., Vol 73, 1990, p 1489-1498

74 Z.H Xia, Y.H Zhou, Z.Y Mao, and B.L Shang, Fabrication of Fiber-Reinforced Metal Matrix Composites

by Variable Pressure Infiltration, Metall Trans B, Vol 23, 1992, p 295-302

75 A Mortensen, V.J Michaud, and M.C Flemings, Pressure-Infiltration Processing of Reinforced Aluminum,

JOM, Vol 45, 1993, p 36-43

76 H You, M.G Bader, Z Zhang, S Fox, and H.M Flower, Heat Flow Analysis of the Squeeze Casting of

Metal-Matrix Composites, Compos Manuf., Vol 5, 1994, p 105-112

77 S Long, Z Zhang, and H.M Flower, Infiltration and Wetting of Liquid Infiltration of Unidirectional Fiber

Arrays by Squeeze Casting, Acta Mater., Vol 42, 1994, p 1389-1397

78 T.R Jonas, J.A Cornie, and K.C Russell, Infiltration and Wetting of Alumina Particulate by Aluminum

and Aluminum-Magnesium Alloys, Metall Trans A, Vol 26, 1995, p 1491-1497

79 S Long, Z Zhang, and H.M Flower, Characterization of Liquid Metal Infiltration of a Chopped Fiber

Preform Aided by External Pressure, Parts I-III, Acta Mater., Vol 43, 1995, p 3489-3509; Vol 44, 1996, p

4233-4240

80 T Yamauchi and Y Nishida, Consideration on Suitable Infiltration Conditions for Molten Metal into

Fibrous Preforms, J Jpn Inst Light Met., Vol 45, 1995, p 409-414

81 J.-H Ahn, N Terao, and A Berghezan, Experimental Factors on Wetting and Infiltration Fronts in Metals,

Scr Mater., Vol 22, 1988, p 793-796

82 J.-H Ahn and A Berghezan, Scanning Electron Microscopy of Liquid Metal Infiltration of Capillaries,

Mater Sci Technol., Vol 7, 1991, p 643-648

83 R.P Courier and S.M Valone, Time-Dependent Solution to the Tai-Chou Chemical Vapor Infiltration

Model, J Am Ceram Soc., Vol 73, 1990, p 1758-1759

84 T.L Starr and A.W Smith, 3-D Modeling of Forced-Flow Thermal-Gradient CVI for Ceramic Composite

Trang 26

Fabrication in Chemical Vapor Deposition on Refractory Metals and Ceramics, T.M Besmann and B.M

Gallois, Ed., Mater Res Soc., 1990, p 55-60

85 T.R Fletcher, J.A Cornie, and K.C Russell, Capacitance Technique for Studying Pressure Infiltration,

86 R Asthana and P.K Rohatgi, Melt Infiltration of Silicon Carbide Compacts, Z Metallkd., Vol 83, 1992, p

89 H Nakanishi, Y Tsunekawa, N Mohri, and I Niimi, Ultrasonic Infiltration in Alumina Particle/Molten

Aluminum System, J Jpn Inst Light Met., Vol 43, 1993, p 14-19

90 Y Tsunekawa, H Nakanishi, M Okumiya, and N Mohri, Application of Ultrasonic Vibration to Molten

Aluminum Infiltration, Key Eng Mater., Vol 104-107, 1995, p 215-224

Infiltration

Claus G Goetzel, Stanford University, and Joanna Groza, University of California, Davis

References

1 P Schwarzkopf, The Mechanism of Infiltration, Symposium on Powder Metallurgy, 1954, Special Report

No 58, The Iron and Steel Institute, London, 1956, p 55-58

2 K.A Semlak, C.W Spencer, and F.N Rhines, Rate of Capillary Rise of Liquid Metal in a High Melting

Metal Powder Compact, Trans AIME, Vol 209, 1957, p 63-64

3 K.A Semlak and F.N Rhines, The Rate of Infiltration in Metals, Trans AIME, Vol 212, 1958, p 325-331

4 W.D Jones, Fundamental Principles of Powder Metallurgy, Edward Arnold, 1960, p 505-512

5 C.G Goetzel and A.J Shaler, Mechanism of Infiltration of Porous Powder Metallurgy Parts, J Met., Vol 16

9 C.G Goetzel, Infiltration Metallurgy, Research, Vol 4 (No 12), 1951, p 555-561

11 G Matsumura, Stress Infiltration in Two-Phase Alloys, P1anseeberi Pulvermetall., Vol 8 (No 3), 1960, p

110-118

12 W.A Kaysser, S Takajo, and G Petzow, Skeleton Dissolution and Skeleton Formation During Liquid

Phase Sintering of Fe-Cu, Modern Development in Powder Metallurgy, Vol 12, Metal Powder Industries

Federation, 1981, p 473-482

13 H.W Lavendel and C.G Goetzel, Recent Advances in Infiltrated Titanium Carbides, High Temperature Materials, R.F Hehemann and G.M Ault, Ed., John Wiley & Sons, 1959, p 140-154

14 G Langford, High Speed Steel Made by Liquid Infiltration, Mater Sci Eng., Vol 28, 1977, p 275-284

15 C.G Goetzel, Treatise on Powder Metallurgy, Vol 2, Interscience, 1950, p 196

16 K Schröter, Border Regions of Metallography, Z Metallkd., Vol 23 (No 7), 1931, p 197-201

Trang 27