Volume 07 - Powder Metal Technologies and Applications Part 10 potx

Based on powder packing density, temperature, bed pressure, and ideal gas laws, parts per million by weight is given by: Loss of part dimensional control in large compacts greater than a

Trang 1

of the available volume inside the container, a considerable amount of shrinkage occurs The science/art of designing a container must account for the packing density and the symmetry or lack thereof to achieve an acceptable part During the initial production of a particular component, preproduction trials and/or iterations of the full-size shape may be necessary

to determine the shrinkages empirically However, this iterative approach is frequently costly and time consuming

Through the years, HIP P/M part manufacturers have employed engineering intuition and previous experience to develop the starting can design At this time, other approaches are being developed and used, namely, empirical and continuum mechanics/finite element modeling Some of these are briefly described here

Empirical Models. A large percentage of the HIP P/M compacts produced are either simple or hollow cylinders An empirical model for these shapes was developed several years ago (Ref 26) by analyzing dimensional data from before and after HIP for a variety of cylindrical shapes To eliminate the effects of the HIP cycle, alloy systems, and container thickness, the analysis focuses on cylinders made from nickel-base alloys consolidated in similar HIP cycles with a certain can thickness range Best fit curves were generated for axial and radial shrinkage as a function of aspect ratio (length to diameter) and surface area ratio (area of cylindrical component to area of the lateral ends) as shown in Fig 13 Based on these data, a computer program was generated to provide either the starting container dimensions to make a finished near-net shape or predict post-HIP dimensions given a specified starting container (Ref 26)

Fig 13 Normalized shrinkage on solid cylinders (normalized shrinkage = actual shrinkage/isotropic shrinkage)

Source: Ref 20

Engineering Models. One promising approach to perform a direct process simulation via the use of computer models

is to combine a constitutive model of continuum mechanics equations solved by a finite element computational method Several approaches ranging from simple plastic to compressible, viscoplastic constitutive models have been investigated

as described in the article "Principles and Process Modeling of Higher-Density Consolidation" in this Volume Even though some of these mathematical models are utilized in production, none has matured into a reliable modeling system for arbitrary geometries and HIP cycles

Container Fabrication

Tooling and Container Component Fabrication. Once the design has been established, the metal container is fabricated This is not a trivial step because the container must be producible in an economical fashion or the finished part cannot be manufactured The most economical and easily formed container material is low-carbon steel; however, other materials (e.g., stainless steel, nickel alloys, titanium, etc.) can also be used The process is constrained by existing metalforming techniques (e.g., metal spinning, hydroforming, stamping, hand forming, casting, machining, etc.) with each having its inherent advantages and limitations

Tooling for the HIP P/M process refers to that which may be used to fabricate the container components Usually, the quantity of parts to be made dictates the precision and cost committed to the tooling for HIP P/M containers Large numbers of parts (1000 or more) would employ stamped containers Anything less than this would be determined on a case-by-case basis Cost of tooling must be amortized over the quantity of parts produced, so more expensive and more

Trang 2

precise tooling can be cost effective only for sizable production runs Because HIP P/M is most often used as a near-net shape process with small quantities of parts, container tooling is not made to be as precise as other processes where net shape is important

Cleaning. Contamination of encapsulated powders will result, unless dirt, oxides, metalworking lubricants, and rust

preventatives used to fabricate container components are removed See Surface Engineering, Volume 5 of the ASM

Handbook for cleaning procedures applicable to various metals Proper cleaning, storage, and handling procedures

immediately prior to any welding operation are necessary to prevent dirt entrapment or contamination on can surfaces

Powder metallurgy alloys that are particularly sensitive to contamination (titanium, nickel-base alloys, and refractory metals) require controlled humidity and stringent cleanliness for final can preparation, assembly, welding, and filling Electropolishing of stainless steel container components and nonchlorinated solvent cleaning (usually acetone, methylethylketone, or methanol) of titanium container components represent typical cleaning processes for specialized applications Carbon steel sheet metal containers should be supported carefully to prevent distortion during welding Similar precautions are recommended for the outer sheet metal container used in a ceramic mold process

Welding. A matched-weld-lip container configuration is designed to promote directional plane front solidification with good liquid metal feeding in the solidifying weld metal Certain oxides (iron, nickel, and copper) can be reduced at high temperature in a high-pressure argon environment This process may produce leaky containers during HIP if oxides extend through weld metal or container wall materials Use of stainless steel filler metal for carbon steel container repairs

is recommended because chromium oxides essentially are stable under processing conditions up to 1200 °C (2200 °F) in argon Gas tungsten arc welding of nickel containers with stainless steel filler metal is also advised

Containers for loose powder are assembled, welded, leak tested, and filled in sequence Containers with interior spacers (mandrels), powder/solid composites (e.g., clad components), or precompacted and sintered P/M compacts can be filled with at least one cover removed This procedure results in an extensive assembly weld area that cannot be leak tested in the vacuum mode because of the slow response time of helium through the interior of the filled container Consequently, careful removal of loose powder from the weld area is necessary and use of precision and reproducible (preferably machine) weld techniques is required to prevent leakage Leak tightness of HIP containers is a major process consideration

Electron beam, gas tungsten arc welding, and stick welding are used for final container assembly Argon dry box and electron beam welding are used for titanium alloy containers because nonoxidizing conditions are required Gas tungsten arc welding with and without filler is used for carbon steel and stainless steel containers Carbon steel containers may require a final reduction anneal after weld assembly, and some clad parts may need to be preheated prior to welding because of substrate material considerations or section size differences Because weld metal is essentially a solidified casting, shrinkage and gas porosity are the fundamental causes of leakage at welds

Leak Testing. Containerized HIP of metal powders can only be achieved successfully with leak-free containers Location of leaks in a fully assembled container by use of valid leak-testing procedures and subsequent repair are fundamental requirements of HIP P/M technology Leak detection is based on characteristics of helium and argon flow through small capillaries when compared at 1 atm (0.1 MPa) and 1000 atm (100 MPa) total pressure Flow characteristics

of a cylindrical capillary have been described by Guthrie and Wakerling (Ref 27):

where Q is the flow rate, cgs units; P1 and P2 are the exterior and interior pressure, cgs units; C1, C2, C3, and C4 are

constants; and L is the capillary length, cm For P2 = 0 (evacuated container interior) and P1 large:

where is the gas viscosity and D is capillary diameter, both in cgs units This applies strictly in the viscous flow region,

when Reynolds number (Re) is <1200:

Trang 3

Re = DV / < 1200 (Eq 3)

where V is the gas velocity, cgs units; and is the gas density, cgs units

Equation 2 indicates the relationship of container design, manufacturing, and leak testing Leakage flow is proportional to the exterior pressure squared Whereas leak testing is conducted with pressure differences across the container wall of one

to several atmospheres, HIP typically uses 1000 atm (100 MPa) Thus, a leak occurring just below the detectability limit

of a selected method permits leakage flow rates 106 times greater during HIP Consequently, sensitivity of the detection method is of utmost importance

leak-Capillary length (L) can be identified with container wall thickness, and with all other variables being constant, a capillary

leaks ten times faster through a 0.25 mm (0.01 in.) wall than a 2.5 mm (0.1 in.) wall The self-sealing, matched-weld-lip design is advantageous because capillary path length through the weld increases rapidly as weld flanges deform and solid-state bonding occurs The fourth power dependence of leakage flow rate on capillary diameter indicates the necessity for procedures to eliminate weld porosity

Leakage flow rates for helium at 1 and 10 atm and for argon at 1000 atm (100 MPa) for a set of capillary leak sizes and

D4/L parameters that could occur in practice are given in Table 1 This illustrates the major problem in leak testing of HIP

containers: the leakage flow rate of argon through a capillary hole at 1000 atm (100 MPa) process pressure is approximately 105 times greater than the flow rate during a 1 atm (0.1 MPa) leak-testing procedure such as use of the helium mass spectrometer in the vacuum mode (i.e., evacuated container and/or atmospheric helium surrounding container exterior) This flow-rate difference defines a requirement for maximum sensitivity of the leak-testing method that is satisfied only by use of the helium mass spectrometer method in the vacuum mode

Table 1 Leakage rate as determined by capillary (hole) diameter, gas type, and pressure

Helium at

1 MPa (10 atm)

Argon at

100 MPa (1000 atm) 0.001 3.8 × 10-16 3.8 × 10-14 3.3 × 10-11

0.01 3.8 × 10-12 3.8 × 10-10 3.3 × 10-7

0.1 3.8 × 10-8 3.8 × 10-6 3.3 × 10-3

1.0 3.8 × 10-4 3.8 × 10-2 3.3 × 10-1

Note: Leakage rate is inversely proportional to the capillary length; that is, if the capillary length is twice as long, the leakage rate will

be half as much For example: if D = 0.1 m helium pressure = 1 MPa, and L = 0.3 cm, then leakage rate = (3.8 × 10-6/3) cm3/s = 1.3

Trang 4

Table 2 Total argon leakage flow at standard temperature and pressure

Capillary (hole) length L = 0.1 cm

Note: Leakage flow is inversely proportional to capillary length; that is, if the capillary length is twice as long, the leakage flow will

be half as much For example, if D = 0.1 m, argon pressure = 100 MPa (1000 atm), leakage time = 1000 s, and L = 0.3 cm, then

leakage flow = 3.8 × 100/3cm3 = 1.3 × 100cm3 Source: Ref 22

Fig 14 Leak testing setup Acceptance criterion: Q (flow rate) <10-9 standard cm 3 /s (a) Test piece evacuated and hooded with helium atmosphere to determine overall leakage rate (b) Test piece evacuated; helium jet probe used to locate leak Source: Ref 22

Fig 15 Argon contamination level versus total leakage flow for various compact sizes Source: Ref 22

Contained argon, although compressed during early powder densification stages in the HIP cycle, can limit end-point densification by "pressure balance" within small remaining pores Regrowth of pores in subsequent heat treating operations, with related adverse effects on properties, can occur at levels as low as 0.1 mL/m3 (0.1 ppm) for tool steels and 1 to 5 mL/m3 (1 to 5 ppm) for superalloys Leaks representing argon contamination at the 10 to 100 mL/m3 (10 to 100 ppm) level generally prevent full densification, and larger leaks usually result in partial or no HIP densification

Compact Manufacture

Trang 5

Loading. Filling of powder into the hermetically sealed, preshaped metal container can be performed in air, under inert gas, or under vacuum conditions, with the latter to aid in the removal of adsorbed gases In some cases, powder is still loaded in open air as it was 25 years ago; however, most processing today is containerized to protect the product and prevent inhalation of the metal powder by operators (Fig 16) Advanced filling systems have been developed to ensure clean, dry handling of powder for critical aerospace applications Magnetic particle separation, screening, outgassing, and settling have been incorporated into these systems In a production operation, there is a need for more sophisticated load stations that are automated to achieve maximum productivity These are usually enclosed systems capable of operating with inert gas or vacuum conditions inside the container and the system Some loading facilities are also capable of hot dynamic outgassing during the filling operation If effective, this type of load station will prevent the need for subsequent outgassing once the compact is filled Figure 17 illustrates a commercial degassing and capsule filling station

Fig 16 Illustration of Modeen loading station

Trang 6

The loading process must be performed in a manner to meet the desired packing density of the powder to obtain the proper post-HIP shape Vibration to settle powder (i.e., packing) is usually employed to get the uniform distribution of powder throughout the inside volume of the container This is important for good shape definition and for reproducibility among parts of the same configuration It is also possible to load powder into the container first, and then by use of a large-amplitude low-cycle vibration pack the powder in the compact This process is sometimes called "thumping." Organic materials such as rubber tubing are not recommended in the powder flow path as they are an obvious contamination source

Particular attention should be paid to completion of loading Figure 18 illustrates recommended and poor container filling practice An incompletely filled container results in loss of shape control and may result in collapse and tearing of the container under external process pressure Compacted powder in the fill tube provides integral contiguous material for testing and evaluation of the part

Fig 18 Container filling practices (a) Poor practice (b) Recommended practice Source: Ref 22

Outgassing The functional requirement of encapsulated powder vacuum/hot outgassing is to remove the atmosphere and water vapor (free and absorbed) from the packed powder bed to prevent formation of particle surface oxide and nitride films, which reduce workability and/or mechanical properties of the subsequent consolidated product Behavior of powder in a heating/vacuum cycle (at temperatures up to approximately 400 °C, or 750 °F), for the purpose of defining process specifications, can be determined by thermogravimetry, combined with limited range mass spectrometry techniques Vacuum outgassing does not remove gas entrapped in hollow powder particles originating from inert gas atmosphere atomization operations Evacuation time for a packed powder bed can be estimated using viscous and molecular flow concepts Elevated temperature is used to raise gas pressure within a bed and to promote desorption of water vapor Packed metal powder beds are poor thermal conductors; therefore, an excessively high heating rate and temperature gradient in the compact during outgassing can result in redistribution of gas by chemisorption and reaction in the outer zone before all the gas is pumped out of the bed This can occur because the center of the bed evolves gas at

"low" temperature, which diffuses and reacts with the outer "high"-temperature portion of the bed before it leaves the compact

The required practical end point for degassing a powder-filled hot isostatic pressing container can be estimated from the residual bed pressure (assuming air composition), which contributes oxygen and nitrogen levels ten times less than the

Fig 17 Commercial degassing and capsule filling station

Trang 7

base level of the powder Based on powder packing density, temperature, bed pressure, and ideal gas laws, parts per million by weight is given by:

Loss of part dimensional control in large compacts (greater than about 25 kg, or 55 lb) also can occur during degassing, because the sheet metal container heats faster than the contained packed powder New empty space is created inside the can at the bottom, into which powder flows from top areas An oversize diameter at the bottom and uneven top geometry result from this type of powder movement during can heating without applied pressure This particular problem also can occur in hot-loading HIP operations

At the completion of the outgassing cycle, the fill/outgas stem is torch heated to approximately 982 °C (1800 °F) and sealed by use of a crimping tool This leaves the powder-filled container sealed under vacuum and ready for consolidation

by HIP Loss of powder during evacuation and degassing (after can filling) can be prevented by inserting a stainless steel wool plug or a metal plug and partial crimping (Fig 19)

Fig 19 Insertion of plugs to prevent loss of powder during evacuation Source: Ref 22

Trang 8

Consolidation by HIP. The HIP of powder compacts is performed to consolidate the powder metal by plastic deformation to 100% of theoretical density Details of the various HIP systems and cycles are discussed in previous sections of this article In a cold-loading system (most commonly used for near-net-shape work), powder-filled compacts are fixtured and loaded into the autoclave (HIP) vessel Pressure and temperature are increased at preprogrammed rates until the desired HIP hold parameters are reached At the completion of the hold, the vessel is depressurized and the furnace is turned off The fully dense parts are removed from the fixturing and sent to the next step in the process

The upper size limitation for densification of encapsulated parts is governed primarily by the processing unit uniform temperature working zone diameter and length Tool steel billets approximately 60 cm (24 in.) in diameter by 300 cm (120 in.) long and larger have been fully densified by HIP Nickel-base superalloy P/M turbine disks greater than 1 m (3.3 ft) in diameter have been successfully densified For sheet metal encapsulation of P/M parts weighing more than approximately 20 kg (44 lb), attachment of handling lugs is recommended For large-diameter parts (greater than 0.5 m,

or 1.6 ft, in diameter) and weights greater than 100 kg (220 lb), sheet metal bending stresses due to enclosed powder weight must be considered Careful consideration must be given to the support of large parts in HIP tooling to prevent distortion during heating prior to complete densification For small net-shape parts (1 to 1000 g, or 0.03 to 35 oz), particularly with thin sections, tooling that permits separate setting of each part is required

Production of small (less than 10 kg, or 22 lb) net-shape parts by the HIP P/M process is not generally economical because it is labor intensive and the container fabrication costs are high This applies particularly to the manufacturing of individual tools from P/M tool steels Exceptions include experimental parts and manufacture of specialty parts in P/M refractory metals, composites, and precious metals, where metal cost is a controlling factor Small net-shape parts (less than 0.5 kg or 1 lb) are best manufactured by containerless HIP, particularly for tool materials, provided satisfactory process procedures can be developed

Postconsolidation Processing. After HIP consolidation, the compact fill stems are removed, and the components are dimensioned The material in the fill stem is checked for density and microstructure Failure to meet set criteria for any of these characteristics is cause for rejection of the compact Many parts are further processed through heat treatment, container removal via chemical dissolution or machining, NDT, and mechanical testing prior to certification and shipment

to the end user

Because the HIP P/M process generally provides a near-net shape, the powder part manufacturer usually supplies a semifinished product This could include material in any one or a combination of the following conditions:

• As-HIP

• HIP plus heat treated

• Rough machined

• NDT qualified

• A preform for thermomechanical processing

• HIP plus thermomechanically processed

The end users determine in what condition they want to receive their parts and specify all requirements such as dimensions, surface, thermomechanical history, properties, and so forth

References cited in this section

1 H.V Atkinson and B.A Rickinson, Hot Isostatic Pressing, 10 P Publishing, 1991

11 R.E Smelser, J.F Zarzour, J Xu, and J.R.L Trasorras, On the Modeling of Near-Net Shape Hot Isostatic Pressing AMD, Mechanics in Materials Processing and Manufacturing, Vol 194, ASME, 1994, p 213-237

17 F.S Biancaniello, J.J Conway, P.I Espina, G.E Mattingly, and S.D Ridder, Particle Size Measurement of Inert Gas Atomized Powder, Mater Sci Eng A, Vol 124, 1990, p 9

18 P Loewenstein, Superclean Superalloy Powders, Met Powder Rep., Vol 36 (No 2), Feb 1981, p 59-64

19 U.S Patent No 4,078,873, 1978

20 J.J Conway, F.J Rizzo, and C.K Nickel, Advances in the Manufacturing of Powder Metallurgy (P/M)

Trang 9

Parts by Hot Isostatic Pressing, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 27-32

21 J.J Conway and J.H Moll, Current Status of Powder Metallurgy Near Net Shapes by Hot Isostatic Pressing, Int Third Conf Near Net Shape Manufacturing (Pittsburgh), ASM International, 27-29 Sept 1993, p 125-

131

22 Product literature and data, Industrial Materials Technology, Inc

23 U.S Patent No 3,622,313, Nov 1971

24 C.F Yolton and J.H Moll, Powder Metallurgy (P/M) Near-Net Shape Titanium Components from Prealloyed Powder, Titanium 1986 Products and Applications, Vol II, Ohio Titanium Development Association, 1987, p 783-800

25 G.S Garibov, V.N Samarov, and V.I Geigin, Powder Metallurgy Industry, Economics, and Organization

of Production, Sov Powder Metall., Vol 18 (No 2), July 1979, p 136-140

26 J.J Conway, "Final Shape Prediction of Hot Isostatic Pressed Powder Metallurgy (P/M) Compacts," MSE

298 Masters Project, University of Pittsburgh, 21 Aug 1990

27 A Guthrie and R.K Wakerling, Vacuum Equipment and Techniques, 1949, p 191

Hot Isostatic Pressing of Metal Powders

J.J Conway and F.J Rizzo, Crucible Compaction Metals

Applications

The ability of HIP to produce near-net shapes has been a primary impetus behind the development of HIP P/M parts Conventional manufacturing methods for materials with high alloy content have low process yields and typically utilize only 10 to 30% of the material purchased in the final product; the remainder becomes scrap during machining Hot isostatic pressing to near-net shape improves material utilization significantly during part manufacturing and finish machining A hot isostatically pressed near-net shape part normally loses only 10 to 20% during final machining The inability to provide nondestructive inspection of complex near-net-shape parts for certification has somewhat inhibited application of this technology, particularly for turbine engine applications

High-Speed Tool Steels. The development of gas-atomized prealloyed steel powders in the 1960s (Ref 3) led to HIP P/M tool steels This represented the first production application of HIP for a relatively low-cost material Hot isostatic pressing improves the microstructure of tool steels by preserving the fine grain size and carbide distribution present in the atomized powder through the consolidation process Increased homogeneity of the fine carbides throughout the material is

an added benefit Superior tool properties result from the improved microstructure Shape stability during subsequent heat treatment is superior in HIP material Grindability, wear resistance, and uniformity of hardness also are improved Additionally, cutting performance of high-speed tool steels is improved by this processing treatment, due to the increased toughness related to fine austenite grain size New high-alloy-content steels with enhanced material properties can be produced High-speed tool steels are generally consolidated in billet form A HIP high-speed steel compact is shown in Fig 20

Trang 10

Nickel-Base Superalloys. Starting with development in the early 1970s, nickel-base superalloys have evolved into one of the best applications for the P/M HIP technology More than 5000 tons (4545 metric tons) of superalloy components are currently operating in commercial and military aircraft turbine engines Hot isostatic pressing of forging preforms represents a significant portion of the current production, but there are approximately 100,000 as-HIP parts in service as well The use of HIP P/M consolidation for superalloys is economically attractive because of its near-net-shape capabilities High-alloy-content superalloys can be produced with attractive properties Superalloys strengthened by a large volume fraction of second-phase ' undergo severe segregation during ingot solidification Such ingots would be virtually unworkable by conventional hot-working techniques for large-size parts The division of the melt into small powder particles during atomization eliminates macrosegregation, and microsegregation is reduced because of high cooling rates during particle solidification Hot isostatic pressing of these powders produces a homogeneous microstructure that improves mechanical properties and hot workability

Superalloy powders are typically made by inert gas atomization or REP Care must be taken in processing to avoid the presence of stable nonmetallic compounds on the surface of the powder particles because they can be detrimental to the properties of consolidated products The article "Powder Metallurgy Superalloys" in this Volume discusses the properties of many nickel-base superalloys made via the HIP P/M process A comparison of HIP properties with other forms is given in Table 3

Table 3 Heat treatments, grain size, and tensile properties of René 95 forms

Heat

treatment/property

Extruded and forged(a)

Hot isostatic pressing(b)

Cast and wrought(c)

(a) AC, air cooled Processing: -150 mesh powder, extruded at 1070 °C (1900 °F) to a reduction of 7 to 1 in

area, isothermally forged at 1100 °C (2012 °F) to 80% height reduction

(b) Processing: -150 mesh powder, HIP processed at 1120 °C (2050 °F) at 100 MPa (15 ksi) for 3 h

(c) Processing: cross-rolled plate, heat treated at 1218 °C (2225 °F) for 1 h

Fig 20 Large-sized cylindrical

high-speed steel billet Courtesy of Crucible

Materials Research Center

Trang 11

Heat treatment after HIP can have significant effects on material properties as shown in Table 4 Material response to post-HIP treatment depends on the processing conditions Near-net-shape parts also may be subject to distortion during post-HIP heat treatment If complex shapes are required, the ceramic mold process is suitable, particularly for static parts

If a carbon or stainless steel container is used for powder consolidation, a 0.5 mm (0.02 in.) diffusion zone may surround the part This does not cause a problem in the final part because HIP envelopes usually exceed this dimension Hot isostatic pressing conditions are alloy dependent Processing temperatures may be keyed to the ' solvus temperatures for purposes of grain size control in nickel-base superalloys

Trang 12

Table 4 Mechanical properties of hot isostatically pressed plus conventionally forged Nimonic alloy AP1

Tensile properties(a) Stress rupture(b) Processing

temperature

Size of sample disk

Yield point, 0.2%

offset

Ultimate tensile strength

Notched tensile strength

°C °F mm in

Solution treatment

MPa ksi MPa ksi

h

Elongation,

%

Notch life,

h

cycle fatigue(c), cycles

Trang 13

Oxide-dispersion-strengthened superalloys also can be consolidated by HIP Prior to processing, alloy powders, additives, and oxide dispersoids are put in a high-attrition ball mill and mechanically alloyed This ensures fine grain size and uniform oxide distribution throughout the powder Hot isostatic pressing produces fully dense material with these microstructural features maintained

Titanium-Base Alloys. Powder production for titanium and titanium alloys requires special setups because of the reactivity of titanium The hydride/dehydride process is the most common way to make titanium powders, but the particles resulting from this process are not spherical and thus do not work well for near-net-shape processing The early method used to make spherical titanium powder was the REP This was later supplanted by PREP to reduce contamination Either of these processes depends on the ability to manufacture bar product of the alloy being made into powder In the late 1980s, an inert-gas-atomizing technique was developed for titanium and its alloys (Ref 30) By the use

of inert atmosphere or vacuum induction skull melting, the titanium alloy is brought to the molten state The liquid is then poured through a metallic nozzle into a high-pressure gas stream The metal breaks up and resolidifies as spherical titanium particles The powder is collected in a cyclone system designed to cool the powder to prevent sintering

There are any number of applications for titanium and titanium alloy powders In the late 1970s and through the 1980s, the Air Force Materials Laboratory supported many programs to develop near-net shapes for military uses (Ref 31) For many reasons, this work never resulted in an ongoing production process, even though there is still some experimental work being performed currently All of the meaningful earlier work was conducted with PREP powder When the gas-atomized powder became available, it was used for all subsequent activities At that time, the emphasis changed to applications needing titanium aluminide powders Because these can be easily made by the skull-melting/gas-atomization process, the bulk of the experimental work is currently being performed in this area The powders are now being used to manufacture metal-matrix composites and intermetallic-matrix composites The advantages of these products are their light weight, high strength, oxidation resistance, and creep resistance at high temperatures

Cemented Carbides. Tungsten-carbide/cobalt tools are the premier example of containerless HIP to achieve full density by removing residual porosity Superior transverse rupture strength results from HIP The wear performance of cutting tools at high speeds is not significantly improved, however, because this behavior is governed by the hardness of the material rather than by its fracture properties Low cobalt content (3%) alloys can be produced by HIP to give enough toughness for use in drawing dies

Fully dense cemented carbide can be finished to give a perfectly smooth surface, which is required for high-quality rolls, dies, mandrels, and extrusion tools Generally tungsten-carbide/cobalt tool materials are manufactured by CIP and sintering of blended powders, followed by HIP Typical conditions for HIP are 1290 °C (2350 °F) at 100 MPa (15 ksi) for

1 h Cemented carbide parts produced using HIP are shown in Fig 21

Refractory Metals. Consolidation of refractory metals by HIP is a two-step

process Processing these materials to net and near-net shape promotes conservation of these critical resources Niobium alloy C-103 (Nb-10Hf-1Ti-5Zr) has been successfully hot isostatically pressed using a duplex cycle Hydride/dehydride and PREP powders are consolidated in a plain carbon steel container filled with powder at 1260 °C (2300 °F) at 100 MPa (15 ksi) for 3 h The container is then removed in a nitric acid solution and further chemically milled in a nitric-hydrofluoric acid solution to remove the alloy/container interaction layer The material is finished in a HIP step at 1590 °C (2900 °F) at

100 MPa (15 ksi) for 3 h to a final density in excess of 99% of theoretical Room-temperature and high-temperature (1650 °C, or 3000 °F) tensile strength and ductility properties compare favorably to wrought alloy properties The ductile/brittle transition temperature is higher (-18 °C versus 160 °C, or 0 °F versus 320 °F, for standard products) in the HIP material due to increased oxygen content Gas content of the hydride/dehydride material results in poorer weldability than the PREP powder Hydrogen embrittlement also occurs in the hydride/dehydride alloy C-103 Vacuum baking at 870 °C (1600 °F) for 2 h eliminates embrittlement, and the alloy will fail in a ductile manner in tensile and Charpy tests

Near-net shape forward bowls manufactured by consolidation of C-103 in the duplex HIP cycle are shown in Fig 22 The diameter of the bowls was within 0.13 mm (0.005 in.) of final dimensions The P/M net shape weighed 0.8 kg (1.8 lb) This, compared with rough forging weighing 1.7 kg (3.8 lb) and a final part weighing 0.6 kg (1.4 lb), illustrates the material savings achieved by HIP to near-net shape

Fig 21 Tungsten-carbide/cobalt

parts produced by HIP Source:

Ref 22

Trang 14

Included is a provision for parts to be low-temperature HIP to a closed porosity condition, decanned, and re-HIP usually at higher temperatures This option can be employed when the powder/container integration (melting, alloying, contamination, etc.) is unacceptable at the preferred higher HIP temperature This technique has been used, for example, for niobium alloys that are initially hot isostatically pressed at 1205

°C (2200 °F) in low-carbon-steel containers, decanned, and re-HIP at 1595 °C (2900

°F) to circumvent an iron-niobium eutectic reaction at 1360 °C (2480 °F)

Stainless Steels. One of the most prominent applications of the HIP P/M technology

is in the area of stainless steels Both duplex and austenitic steels have been used extensively as P/M near-net shapes in the oil and gas and petrochemical industries For example, valve bodies, fittings, and large complex manifolds for piping systems have been successfully produced in a cost-effective manner via HIP processing Figure 23 (Ref 32) shows some of the typical fittings that have been made from 254 SMO material Figure 24 (Ref 32) is a valve body that weighs more than 2 tons and was made from an austenitic stainless steel Large manifolds with integral outlets hot isostatically pressed from a superduplex stainless steel have also been put in service in an offshore oil rig in the North Sea (Ref 32) In addition to the other benefits of a HIP P/M approach, the manifold can be fabricated in far less time and avoid costly welding processes An analysis of the cost factors showed a greater than 20% savings over a similar manifold produced from fabricated cast and wrought components (Ref 32)

Fig 23 Tees for underwater applications in the offshore industry hot isostatically pressed in 254 SMO grade

Weight: 155 kg/pc

Fig 24 Hot isostatically pressed valve body in austenitic stainless steel Weight: 2 t

3 C.S Boyer, History: Development of a HIP Apparatus to Fulfill a Commercial Need, Hot Isostatic Pressing Conf., ASM International, 20-22 May 1996

Fig 22 Niobium forward

bowls hot isostatically

pressed to shape Source:

Ref 22

Trang 15

28 S Reichman and D.S Chang, Superalloys II, C.T Sims, N.S Stoloff, and W.C Hagel, Ed., John Wiley & Sons, 1987, p 459

29 R.V Miner and S Gayda, Int J Fatigue, Vol 6 (No 3), 1984, p 189

30 U.S Patent No 4,544,404

31 V Peterson, V Chandhok, and C Kelto, Hot Isostatic Pressing of Large Titanium Shapes, Powder Metallurgy of Titanium Alloys, F Froes and J Smugeresky, Ed., AIME, 1980, p 251

32 C.G Hjorth and H Eriksson, New Areas for HIPing Components for the Offshore and Demanding Industries, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May

1996, p 33-38

Interface/Diffusion Bonding

Not only can HIP be used to consolidate loose powder, it can also be used to create a component of multiple bonded materials Diffusion bonding by HIP can be performed on solid-to-solid, powder-to-solid, and in some cases, powder-to-powder surfaces As with powder/metal container combinations, material compatibility must be evaluated to ensure no low-temperature melting reactions occur at the HIP temperature If this does occur, interlayers can be used to alleviate this problem

HIP Diffusion Bonding versus Other Joining Processes. As stated previously, HIP technology was initially developed as a method to diffusion bond two materials together Table 5 shows various attributes of joining two materials when comparing diffusion bonding with fusion methods (i.e., welding and brazing) The major advantages of diffusion bonding are no melting of the parent metal and therefore no segregation or cracking problems, very little dimensional distortion, and stronger bonds due to the elimination of a low-melting-point filler

Table 5 Diffusion bonding in comparison with other joining processes

Fusion welding Diffusion bonding Brazing

Contacting method Autogenous fusion, autogenous

fusion and pressure, pressure and autogenous fusion

Pressure (no fusion) Contact fusion, contact fusion

and pressure, pressure and contact fusion

Temperature Melting point of parent metal 0.5-0.7 of melting point of parent

metal

Somewhat above melting point

of braze

Surface preparation Less exacting Careful Less exacting

Materials Metals, alloys Metals, alloys, nonmetals Metals, alloys, nonmetals

Joint formation Gradual Simultaneous Simultaneous, gradual

Porosity Shrinkage, blowholes None Blowholes, shrinkage, diffusion

Overlapping with

heat treatment

Principle types of Butt, lap Flat (butt, lap, tapered plug in Butt, lap

Trang 16

joint socket, between cylinders,

spherical, curvilinear)

Joining in

hard-to-reach places

Corrosion resistance Satisfactory Fairly high Low

Strength Close to that of parent metal That of parent metal That of braze

Air pollution and

• Powder and porous bodies can be simultaneously densified to a substrate with HIP diffusion bonding

Encapsulation Methods. As with consolidating P/M compacts, components for HIP diffusion bonding must be encapsulated to ensure a differential pressure exists to create the driving force for bonding One method is to simply weld the contact area between the two parts Another is to seal only the contact area with a container component Yet another is

to encapsulate part or all of the substrate Figure 25 shows the steps used to HIP diffusion bond a powder material with a solid substrate material (Ref 21)

Fig 25 HIP diffusion bonding of powder to solid Source: Ref 21

Trang 17

HIP Parameters. The choice of HIP parameters is usually based on metallurgical and economic consideration Diffusion bonding is typically enhanced by increasing temperature and pressure The temperature will generally be 50 to 70% of the melting point of the lowest-temperature material in the system The pressure shall be sufficient to close up all pores along the bond line as well as internal pores and pores created by interdiffusional pores (i.e., Kirkendall effect) The time at temperature should be kept to a minimum to decrease cost and potentially avoid any deleterious effects from formation of brittle intermetallics, excessive grain growth, and secondary recrystallization

Use of Interlayers. An interlayer is sometimes used between surfaces to prevent the formation of deleterious brittle compounds and/or alleviate stresses due to thermal expansion mismatch As described previously, interlayers must be compatible with each material that it contacts The thickness must be sufficient enough to accommodate cooling stress and not so thick that the bond strength is decreased by the presence of a thick ductile interlayer A 100 m thick Ni-Cu-

Ni interlayer was successfully used as a carbon diffusion barrier between BG42 tool steel and 17-4 stainless steel and a cobalt-base alloy and 17-4 stainless steel, thus maintaining a martensitic structure up to the interface (Ref 33) Another interlayer application was the use of refractory metal and ceramic interlayers during the fabrication of as-HIP foil of highly alloyed material (e.g., titanium, nickel, and niobium alloys) (Ref 34)

Applications. There have been several applications of bimetallic components that have utilized the HIP diffusion bonding process Examples include:

• Corrosion-resistant alloy 625 clad to the interior of F22 steel (Fig 26)

• Wear/corrosion resistant alloy (MPL-1) clad to 4140 steel (Ref 21)

• Alloy CPM 9V clad on the exterior of 4140 cylinders (Ref 20)

• Twin extrusion barrel internally clad with CPM 10V against 4140 steel (Ref 21)

• CPM 10V clad to low-carbon steel for segmented screws used inside the plastic extrusion barrel (Fig 27)

Fig 26 Low-alloy steel HIP clad with alloy 625 for corrosion resistance

Fig 27 Bimetallic wear-resistant screw segments for the plastic extrusion industry

Trang 18

20 J.J Conway, F.J Rizzo, and C.K Nickel, Advances in the Manufacturing of Powder Metallurgy (P/M) Parts by Hot Isostatic Pressing, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 27-32

131

33 M.A Ashworth, M.H Jacobs, G.R Armstrong, R Freeman, B.A Rickinson, and S King, HIP Diffusion Bonding for Gear Materials, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 275-285

34 A.M Ritter, M.R Jackson, D.N Wemple, P.L Dupree, and J.R Dobbs, Processing of Metal Foil by Direct HIP of Powder, Aeromat '96 (Dayton, OH), 5 June 1996

Future Developments

Applications using HIP technology have evolved from diffusion bonding of dissimilar materials to consolidating encapsulated powder and sealing microporosity in castings Hot isostatic pressing technology is continuing to grow with diversification into new areas These areas include equipment improvements, mechanistic modeling of material undergoing HIP, and new applications of HIP

Refinements of Batch Processing. One equipment refinement that is generating interest is "quick cool" or "HIP quenching." After the HIP cycle hold, furnace cooling on a cold-walled vessel can take several hours with cooling rates of about 100 °C to 200 °C/h depending on the vessel and size of the load By utilizing a flow device (Ref 15) and the introduction of cold gas into the hot gas, the convective cooling is dramatically increased One portion of the gas is forced

to the outside of the thermal barrier for cooling while the other portion is circulating inside To achieve the desired cooling rate, the proportion of the hot and cold gas can be computer controlled The major driver of this technological improvement is to increase productivity, which ultimately increases capacity and decreases costs In addition, there may

be metallurgical enhancements of some materials, thus potentially eliminating some downstream processing steps

HIP Modeling and Microstructure Prediction. As described in the article "Principles and Process Modeling of Higher Density Consolidation" in this Volume, there has been much work devoted in the 1990s (Ref 13) to predicting dimensional changes during hip via continuum mechanics/finite element modeling To predict shrinkage changes, an understanding is needed of the anisotropy of consolidation brought about by the complex interrelationships between the properties of the P/M and container materials as a function of temperature, density, and part geometry With the development of the constitutive equations for the particles and powder aggregates to predict shrinkage, the underlying mathematics now exist to also predict microstructure of the HIP product (Ref 11, 12, 35) With computational power continually increasing at an affordable rate and material property characterization available from hot triaxial compaction tests (Ref 36), the ability to predict grain size (Ref 37) and other microstructural features (Ref 12, 38) may soon be possible

HIP Modeling and Closing Porosity in Spray Formed Billets. To compete with ring-rolled products, there has been some interest in producing large nickel-base superalloy rings via spray forming followed by HIP (Ref 39, 40) For this process, metal is nitrogen-gas-atomized onto a low-carbon steel substrate to form a partially dense preform (typically,

>90%) The resulting microstructure is determined by amount of liquid in the spray before impact and amount of liquid

on the top surface of the deposit As the amount of liquid is increased, an increase in deposit yield is observed (i.e., atomizing into a swamp); however, these slower solidification rates typically lead to a coarser grain size If a finer grain

Trang 19

size is required, the amount of liquid is decreased, but this typically increases the amount of unusable overspray that cannot be recycled due to increased nitrogen content concerns Hot isostatic pressing of the preform increases the density

to nearly 100% density with some interconnected surface porosity present

12 R.D Kissinger, The Densification of Nickel Base Superalloy Powders by Hot Isostatic Pressing, Diss Abstr Int., Vol 49 (No 9), Mar 1989, p 1-39

13 W.B Eisen, Modeling of Hot Isostatic Pressing, Rev Partic Mater., Vol 4, 1996

15 C Bergman, J Westerlund, and F.X Zimmerman, HIP Quench Technology, Hot Isostatic Pressing: Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 87-90

35 J.F Zarzour, J.R.L Trasorras, J Xu, and J.J Conway, Experimental Calibration of a Constitutive Model for Hot Isostatic Pressing (HIP) of Metallic Powders, Advances in Powder Metallurgy and Particulate Materials 1995, Vol 2, M Phillips and J Porter, Ed., Metal Powder Industries Federation, p 5-89

36 H.R Piehler and D.M Watkins, Hot Triaxial Compaction: Initial Results for Aluminum Compacts, Advances in Powder Metallurgy, Vol 1, E.R Andreotti and P.J McGeehan, Ed., Metal Powder Industries Federation, 1990, p 393-398

37 B.A Hann, I Nettleship, and S Schmidt, An Investigation of Microstructural Evolution of PM Alloy N625 During Interrupted Hot Isostatic Pressing (HIP) Cycles, Superalloys 718, 625, 706 and Various Derivatives,

E Loria, Ed., TMS, 1997, p 781-789

38 M.C Somani, N.C Birla, Y.V.R.K Prasad, and V Singh, Deformation Behaviour and Process Modelling

of Hot Isostatically Pressed P/M Alloy Nimonic AP-1 and Its Correlation with Microstructure, Advances in Materials and Processes, IBH Publishing, 1993, p 104-143

39 N Paton, T Cabral, K Bowen, and T Tom, Spraycast-X 718 IN718 Processing Benefits, Superalloys 718,

625, 706 and Various Derivatives, E Loria, Ed., TMS, 1997, p 1-16

40 T.F Zahrah, R Dalal, and R Kissinger, Intelligent HIP Processing of a Spraycast-X Superalloy for Aerospace Applications, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 163-166

References

2 E.S Hodge, Elevated Temperature Compaction of Metals and Ceramics by Gas Pressures, Powder Metall., Vol 7 (No 14), 1964, p 168-201

3 C.S Boyer, History: Development of a HIP Apparatus to Fulfill a Commercial Need, Hot Isostatic Pressing Conf., ASM International, 20-22 May 1996

4 J.E Coyne, W.H Everett, and S.C Jain, Superalloy Powder Engine Components: Controls Employed to Assure High Quality Hardware, Powder Metallurgy Superalloys, Aerospace Materials for the 1980's, Vol 1, Metal Powder Report Publishing, Shrewsbury, England, 18-20 Nov 1980, p 24-1 to 24-27

5 J.H Moll and F.J Rizzo, Production Applications of Rapidly Solidified Tool Steels, Superalloys, Titanium Alloys, and Corrosion-Resistant Alloys, Rapid Solidification Processing Principles and Technologies III, R Mehrabian, Ed., 6-8 Dec 1982, p 686-691

Trang 20

6 W.B Eisen, P/M Tool and High Speed Steel: A Comprehensive Review, Proc of the 5th Int Conf on Advanced Particulate Materials and Processes, Metal Powder Industries Federation, 1997, p 55

7 W Stasko, K.E Pinnow, and R.D Dixon, Particle Metallurgy Cold Work Tool-Steels Containing 3-18% Vanadium, Proc 5th Int Conf Advanced Particulate Materials and Processes, Metal Powder Industries Federation, 1997, p 401

8 C.M Sonsino, Fatigue Design for Powder Metallurgy, PM-90, World Conf Powder Metallurgy, Vol 1, Institute of Materials, 1990, p 42-88

9 R.M German, Powder Metallurgy Science, 2nd ed., Metal Powder Industries Federation, 1994, p 302-340

10 P Hellman, Review of HIP Development, Hot Isostatic Pressing Theories and Application, Centek Publishers, 1988, p 3-18

12 R.D Kissinger, The Densification of Nickel Base Superalloy Powders by Hot Isostatic Pressing, Diss Abstr Int., Vol 49 (No 9), Mar 1989, p 1-39

13 W.B Eisen, Modeling of Hot Isostatic Pressing, Rev Partic Mater., Vol 4, 1996

14 E Artz, M.F Ashby, and K.E Easterling, Practical Applications of Hot Isostatic Pressing Diagrams: Four Case Studies, Metall Trans A, Vol 13, Feb 1983, p 211-221

15 C Bergman, J Westerlund, and F.X Zimmerman, HIP Quench Technology, Hot Isostatic Pressing: Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 87-90

16 D.J Evans and D.R Malley, "Manufacturing Process for Production of Near Net Shapes by Hot Isostatic Pressing of Superalloy Powder," Final Report on AFWAL-TR-83-4022, Air Force Wright Aeronautical Laboratories, June 1983

17 F.S Biancaniello, J.J Conway, P.I Espina, G.E Mattingly, and S.D Ridder, Particle Size Measurement of Inert Gas Atomized Powder, Mater Sci Eng A, Vol 124, 1990, p 9

18 P Loewenstein, Superclean Superalloy Powders, Met Powder Rep., Vol 36 (No 2), Feb 1981, p 59-64

19 U.S Patent No 4,078,873, 1978

20 J.J Conway, F.J Rizzo, and C.K Nickel, Advances in the Manufacturing of Powder Metallurgy (P/M) Parts by Hot Isostatic Pressing, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 27-32

131

23 U.S Patent No 3,622,313, Nov 1971

24 C.F Yolton and J.H Moll, Powder Metallurgy (P/M) Near-Net Shape Titanium Components from Prealloyed Powder, Titanium 1986 Products and Applications, Vol II, Ohio Titanium Development Association, 1987, p 783-800

25 G.S Garibov, V.N Samarov, and V.I Geigin, Powder Metallurgy Industry, Economics, and Organization

of Production, Sov Powder Metall., Vol 18 (No 2), July 1979, p 136-140

26 J.J Conway, "Final Shape Prediction of Hot Isostatic Pressed Powder Metallurgy (P/M) Compacts," MSE

298 Masters Project, University of Pittsburgh, 21 Aug 1990

27 A Guthrie and R.K Wakerling, Vacuum Equipment and Techniques, 1949, p 191

28 S Reichman and D.S Chang, Superalloys II, C.T Sims, N.S Stoloff, and W.C Hagel, Ed., John Wiley & Sons, 1987, p 459

29 R.V Miner and S Gayda, Int J Fatigue, Vol 6 (No 3), 1984, p 189

30 U.S Patent No 4,544,404

31 V Peterson, V Chandhok, and C Kelto, Hot Isostatic Pressing of Large Titanium Shapes, Powder Metallurgy of Titanium Alloys, F Froes and J Smugeresky, Ed., AIME, 1980, p 251

Trang 21

32 C.G Hjorth and H Eriksson, New Areas for HIPing Components for the Offshore and Demanding Industries, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May

1996, p 33-38

33 M.A Ashworth, M.H Jacobs, G.R Armstrong, R Freeman, B.A Rickinson, and S King, HIP Diffusion Bonding for Gear Materials, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 275-285

34 A.M Ritter, M.R Jackson, D.N Wemple, P.L Dupree, and J.R Dobbs, Processing of Metal Foil by Direct HIP of Powder, Aeromat '96 (Dayton, OH), 5 June 1996

35 J.F Zarzour, J.R.L Trasorras, J Xu, and J.J Conway, Experimental Calibration of a Constitutive Model for Hot Isostatic Pressing (HIP) of Metallic Powders, Advances in Powder Metallurgy and Particulate Materials 1995, Vol 2, M Phillips and J Porter, Ed., Metal Powder Industries Federation, p 5-89

36 H.R Piehler and D.M Watkins, Hot Triaxial Compaction: Initial Results for Aluminum Compacts, Advances in Powder Metallurgy, Vol 1, E.R Andreotti and P.J McGeehan, Ed., Metal Powder Industries Federation, 1990, p 393-398

37 B.A Hann, I Nettleship, and S Schmidt, An Investigation of Microstructural Evolution of PM Alloy N625 During Interrupted Hot Isostatic Pressing (HIP) Cycles, Superalloys 718, 625, 706 and Various Derivatives,

E Loria, Ed., TMS, 1997, p 781-789

38 M.C Somani, N.C Birla, Y.V.R.K Prasad, and V Singh, Deformation Behaviour and Process Modelling

of Hot Isostatically Pressed P/M Alloy Nimonic AP-1 and Its Correlation with Microstructure, Advances in Materials and Processes, IBH Publishing, 1993, p 104-143

39 N Paton, T Cabral, K Bowen, and T Tom, Spraycast-X 718 IN718 Processing Benefits, Superalloys 718,

625, 706 and Various Derivatives, E Loria, Ed., TMS, 1997, p 1-16

40 T.F Zahrah, R Dalal, and R Kissinger, Intelligent HIP Processing of a Spraycast-X Superalloy for Aerospace Applications, Hot Isostatic Pressing, Proc Int Conf Hot Isostatic Pressing, ASM International, 20-22 May 1996, p 163-166

B.L Ferguson, Deformation Control Technology, Inc.; P.R Roberts, American Superconductor Corporation

Introduction

EXTRUSION is a relatively recent addition to metalworking as noted in a historical survey of extrusion and the development of the process (Ref 1) Notably, the inventive genius of Alexander Dick and the increasing availability of steels that could withstand higher working temperatures opened the way for the hot extrusion of copper alloys and laid the foundation for modern extrusion Pearson and Parkins (Ref 1) and Lange and Stenger (Ref 2) have written comprehensively on the history, development, application, and mechanics of extrusion; these are recommended texts that provide an excellent background for understanding the process

There are two main types of extrusion mechanisms, (a) direct and (b) indirect or inverted, as shown in Fig 1 In direct extrusion, the ram pushes a workpiece forward through a die, causing a reduction in cross-sectional area of the workpiece Conversely, in indirect extrusion, the workpiece remains stationary relative to the container and there is no friction between the workpiece and container Both methods may be used to extrude metal powders, although direct extrusion is more widely practiced

Trang 22

Fig 1 Basic methods of extrusion (a) Direct extrusion (b) Indirect extrusion

Powder extrusion provides a method to obtain a useful shape or form that may not be readily achieved by other means It has been used to make seamless tubes, wires, and complex sections that would be difficult or impossible to fashion by any other process Pioneering work in the extrusion of metal powders was conducted in the late 1950s to produce controlled ductility in beryllium, dispersions of nuclear fuels and control rod materials, and dispersion-strengthened aluminum (Ref 3)

The extrusion of metal powders occupies a special niche in extrusion technology for many reasons, which include:

• The ability to form shapes by extrusion from materials that are difficult or impossible to process by casting or working

• Improved properties and performance because of microstructural refinement and minimization of segregation that results from powder processing

• The dispersion of one species in another from the extrusion of powder mixtures

• The ability to form wrought structures from powder without the need for sintering or other thermal treatments

• Reduced extrusion pressures and wider temperature and ram velocity ranges for powder extrusion than for extrusion of cast billets

This article concentrates on direct extrusion processing where metal powders undergo plastic deformation, usually at an elevated temperature, to produce a densified and elongated form having structural integrity Three main approaches to extrusion of powders are shown in Fig 2 In the first instance where the principal material is poured loose into the extrusion container, the particle size is normally large Lenel (Ref 4) mentions that this process is used to extrude magnesium alloy pellets (particle size 70 to 450 m) A hot container supplies heat to the pellet charge, and extrusion is performed with no atmosphere protection

Trang 23

Fig 2 Hot extrusion methods for metal powders

The second method, shown in Fig 2, relies on the use of a compacted billet as the extrusion workpiece Precompaction is useful because a workpiece shape that can be handled is produced, supplying a form that is much easier to use and control

in a manufacturing environment than loose powder, and compaction increases the density of the workpiece in comparison with that of loose powder The higher density reduces the ram stroke and decreases the container length needed to produce the required extruded length In this method, a powder that compacts readily is used; this type of powder has particles that are rough and jagged with multiple asperities and ragged protrusions, or a flake form In loose form, such particles have low packing efficiency and a correspondingly low density, for example, 35 to 50% of theoretical However, these particles can be pressed into a partially consolidated billet shape with densities of 70 to 85% of theoretical This form is referred to as a "green" compact, and it has sufficient "green strength" to endure handling Where greater resistance to crumbling during handling is needed, the porous compact may be sintered before extrusion However, sintering is not a required processing step, and many materials are extruded without this additional processing Alternatively, hot pressing may be used instead of cold pressing to produce the extrusion workpiece

A more elaborate approach to powder billet preparation is that shown in the third variation in Fig 2 and in more detail in Fig 3, where powder is first partially densified directly in a can This can may then be evacuated and sealed, as shown in Fig 4, or it may be left open to the atmosphere Canning is employed for the following reasons (Ref 5):

• Isolation of the principal material from the atmosphere and extrusion lubricants (clean extrusion technique)

• Isolation of toxic materials such as beryllium and uranium for safe handling

• Encapsulation of spherical and other difficult-to-compact powders to produce a billet form

• Improved lubricity and metal flow at the die interface by proper selection of the can material

• Isolation of the principal material from the extrusion die and region of highest shear, which is an important consideration for materials with limited ductility

• The ability to position powder and solid components within the can to produce unique and complex

Trang 24

shapes (this is a variation of the filled billet extrusion technique that is discussed below)

Fig 3 Packing of powder in metal can

Fig 4 Evacuation and sealing of powder extrusion billet (a) Billet with evacuation tube leading to vacuum

pump (b) Billet with sealed tube

When purity must be maintained, canning of the powder, including evacuation and sealing as shown in Fig 4, is an essential step Billet preparation for critical applications, such as gas turbine engine components, requires that filling and evacuation be performed with great care and refined practice Procedures may include clean-room container preparation and assembly, total isolation of powder from ambient air, evacuation at slightly elevated temperatures to drive off adsorbed gases from particle surfaces, and leak checking of sealed containers The extruded product must be free from both prior particle boundary decorations and nonmetallic inclusions that degrade mechanical properties, especially fracture toughness and fatigue resistance

Trang 25

With this introduction to the basic powder extrusion processes, it is pertinent to review briefly the mechanics of extrusion and to examine specific extrusion practices for the production of wrought material from powder stock

References

1 C.E Pearson and R.N Parkins, The Extrusion of Metals, 2nd ed., London, Chapman and Hall, 1960

2 K Lange and H Stenger, Extrusion: Process, Machinery, Tooling, American Society for Metals, 1981

3 P Loewenstein, L.R Aronin, and A.L Geary, Powder Metallurgy, W Leszyski, Ed., Interscience, 1961, p 563-583

4 F.V Lenel, Powder Metallurgy Principles and Applications, Metal Powder Industries Federation, 1980

5 P Roberts, Tech Paper MF 76-391, SME, 1976

Note

* Adapted from article by P.R Roberts and B.L Ferguson, "Extrusion of Metal Powders,"

International Materials Reviews, Vol 36 (No 2), 1991, p 62-79 with review by Peter W Lee, The Timken Company and Donald Byrd, Wyman Gordon Forgings

Extrusion of Metal Powders *

Mechanics of Powder Extrusion

Typical pressure curves for direct and indirect extrusion of a conventional billet are shown in Fig 5 Initially, pressure increases linearly with ram displacement as the billet upsets to fill the container and reaches a maximum value as the workpiece begins to flow through the die; this is known as the breakthrough pressure Steady state is achieved as the ram advances For direct extrusion, the pressure falls as the ram stroke continues, reflecting the decreasing frictional resistance

as the contact area between the billet and container decreases For indirect extrusion, the extrusion pressure is fairly constant because there is no relative movement and thus no friction between the billet and container At the end of the stroke, a sharp rise in pressure may occur because of the increasing resistance to radial inflow of the residual billet material This latter effect may be overcome by placing a follower of some disposable material between the billet and ram

to ensure that the billet clears the die

Fig 5 Extrusion pressure as function of ram travel

Trang 26

Extrusion pressure is a function of material flow stress, temperature, friction, strain rates, and deformation Sachs and

Eisbein (Ref 6) proposed an extrusion constant K such that:

where P is the required extrusion pressure, R the extrusion ratio Ai/Af, which is the ratio of initial cross-sectional area to

final cross-sectional area, and K the extrusion constant for the material

The extrusion constant K combines flow stress, friction, and redundant work into one parameter While K is dependent to

some extent on particular extrusion practices, reported extrusion constant values may be broadly applied as an aid to the design of useful extrusions Extrusion constant values for a variety of engineering materials are shown as a function of temperature in Fig 6 These curves were compiled from many runs on various extrusion presses (Ref 7)

Fig 6 Extrusion constant versus temperature for some engineering alloys

It is clear from Fig 6 that K decreases with temperature, and it follows that hot extrusion requires lower extrusion pressure than cold extrusion The effect of temperature on K is similar to the relationship between flow stress and temperature For cast billets, K increases as the ram speed is increased, which reflects the relationship between flow stress

and strain rate Dieter (Ref 8) mentions that a tenfold increase in ram speed results in a 50% increase in the required extrusion pressure, as a general rule for cast/wrought billet stock

In evaluating the dependency of extrusion pressure on temperature and strain rate, adiabatic heating, heat generated by friction, and conductive heat transfer must be taken into account Proper extrusion speed and temperature are usually determined by experience, following general guidelines shown schematically in Fig 7 This figure shows the relationship between pressure and temperature on deformation level and between strain rate and temperature on allowable deformation

Trang 27

level Together these relationships define an allowable range of processing conditions With improved mathematical modeling capabilities, analytical methods may be used to verify selected extrusion conditions by computer simulation

Fig 7 General effects of pressure, strain rate, and temperature on allowable deformation Source: Ref 8

The discussion up to this point has only dealt with extrusion of cast/wrought billets Extrusion of powder is mechanically different from extrusion of cast material First, the billet stock for powder extrusion is usually compacted powder, with a density of 75 to 85% of theoretical, or loose powder, with a density of 55 to 65% of theoretical The presence of porosity and unbonded particulate changes the deformation mechanics significantly from that of cast/wrought material (Ref 3, 4, 5, 9, 10, 11, 12, 13, 14, 15, 16)

Pressure versus ram displacement curves are shown in Fig 8 for the extrusion of cold compacted Al-Mg-Si (as supplied) powder billets at two ram speeds (Ref 13) For comparison, a pressure versus displacement curve for a similar cast material is also shown The initial pressure buildup is considerably different for powder and cast billets Pressure increases linearly for a solid billet as it upsets to fill the container For a powder billet, the pressure rise during the upsetting stage is both nonlinear and more gradual than for a cast billet During this pressure buildup, the powder is being compacted to nearly full density While densification is substantially completed before the onset of extrusion, powder particles remain poorly bonded and mechanical strength is low The material does not attain measurable strength and ductility until it has traversed through the extrusion die and been subjected to a sufficient level of shear deformation Shear deformation results in the formation of metallurgically sound bonds between particles and produces a "wrought" product

Trang 28

Fig 8 Load versus displacement curves for extrusion of Al-0.33Mg-0.12Si (final) alloy powder (containing

alumina) and cast billet Source: Ref 12

In practice, back pressure or resistance must be provided initially to aid densification of the powder in the billet nose before the extrusion Therefore, a solid nose or a pad is placed in front of the powder billet The pad may also provide a measure of lubrication Extrusion of green compacts without such assistance to densification has resulted in gross surface cracking (Ref 9) For loose powder, the pad is incorporated in the container design

Experimental results for extrusion of compacted billets have shown that the extrusion pressure is less dependent on ram speed for powder than for cast material (Ref 9) Therefore, strain rate is of less importance in powder extrusion than in extrusion of cast materials The major reason for this is thought to be the overwhelming influence of porosity and the formation of particle bonds during powder deformation as opposed to the importance of flow stress on extrusion pressure for cast/wrought material However, the breakthrough load for powder extrusion remains a function of strain rate and increases as ram speed increases (Ref 13)

Theoretically, the pressure P needed to extrude a billet through a die has three components:

Trang 29

ratios The ratio a'/b' indicates the relative importance of redundant work, and it is clear that redundant work is much

more significant in powder extrusion than in extrusion of cast stock For powder extrusion, the redundant work term includes the process of forming cold welds between particles, breaking welds, and rewelding, while particle deformation

is occurring Sheppard proposed that the high proportion of redundant work in extrusion of powder contributes to the formation of sound particle bonds and eliminates the need for subsequent sintering (Ref 9, 10, 11, 12, 13, 14, 15) A minimum level of extrusion reduction must be exceeded in order to produce sound bonds, with 9 to 1 being a typical minimum reduction ratio for extrusion of spherical powders

Table 1 Extrusion pressure and reduction ratio relationships based on P = a' + b' ln R

Powder, wt% Extrusion relationship a'/b'

Sheppard and coworkers have extended the upper-bound analysis to study powder extrusion (Ref 9, 10, 11, 12, 13, 14, 15) A typical deformation zone geometry, as defined by the upper-bound method, and the accompanying hodograph are shown in Fig 9 This analytical technique is reviewed thoroughly by Sheppard et al (Ref 9, 11, 12) and Avitzur (Ref 17)

Fig 9 Typical deformation zone and hodograph for upper bound modeling of extrusion of powder billet oa,

initial velocity Vi; oc, final velocity Vf Source: Ref 11

For canned powders, an important consideration is the nature of the flow of material through the die Modeling studies of material flow during extrusion have been conducted using such diverse materials as colored wax (Ref 18), mixtures of chalk, beeswax, and petroleum jelly with added coloring (Ref 19), and Plasticene (Ref 20) Green (Ref 20) has shown that

a good working analogy to hot metal is afforded by such simple substances Sachs and Eisbein (Ref 6) used small cylindrical tin billets that were sectioned longitudinally, with a grid scribed on the flat surface The cylinder halves were held together and forced through conical dies of different included angles The flow patterns were revealed in the

Trang 30

distortion of the grid in a range of partial extrusions, as shown in Fig 10 When the die is flat (180° included angle), a zone of dead metal forms in the front corner of the press container and flow at the periphery of the billet occurs by shear over a conically defined surface (Ref 6, 17) As the included cone angle decreases from the extreme case of the flat die, flow becomes more streamlined and the ogival characteristic of the transverse grid lines becomes less pronounced Conical dies are used for the extrusion of canned powder to avoid the formation of a stagnant flow zone and the associated turbulence and shear experienced with flat dies Conical dies work well with included angles of 90 to 120°, and good streamlined flow is obtained when there is effective lubrication and closely fitting extrusion tooling working in conjunction with the die Canned extrusions have been limited to the fabrication of simple shapes such as circular sections, tubes, flats, and so forth, because of the necessity for a conical approach to the die land Advances in analytical modeling of extrusion and in die-making techniques permit the extrusion of more complex cross sections

Fig 10 Effect on distribution of flow caused by use of dies of different conicity Small scale experiments with

cylindrical tin samples Source: Ref 6

Comparison of Extrusion with Cast and Powder Billets. Comparison of analysis results for cast billets with results for powder billets has led to some fundamental conclusions:

• Extrusion of powder is a continuous process of welding particles, breaking welds, and rewelding as particles are rearranged and deformed Densification is nearly completed in the extrusion container, before actual extrusion However, the achievement of adequate structural integrity is dependent on shear deformation as the compacted powder is forced through the extrusion die

• The presence of porosity and unbonded particles in the extrusion billet produce a deformation zone much different from that of the extrusion of a cast billet The extrusion pressure for powder is much lower than for extrusion of a cast billet of similar composition This results from the development of final properties in the powder extrusion only after the material has traversed through the deformation zone Conversely, a cast billet exhibits higher flow stress before extrusion

• The amount of redundant work is dependent on the surface area/volume ratio of the powder As the powder particle size decreases, this ratio increases, and a greater number of welds are required for bonding Consequently, the amount of redundant work increases

Trang 31

6 G Sachs and W Eisbein, Mitt Deut Mater Pruf Anst., Vol 16, 1931, p 67-96

7 unpublished data, Nuclear Metals, Inc

8 G.E Dieter, Mechanical Metallurgy, 3rd ed., McGraw-Hill, 1986

9 T Sheppard and P.J.M Chare, Powder Metall., Vol 15 (No 29), 1972, p 17-41

10 P.J.M Chare and T Sheppard, Powder Metall., Vol 16 (No 32), 1973, p 437-458

11 P.J.M Chare and T Sheppard, Int J Powder Metall Powder Technol., Vol 10 (No 3), 1974, p 203-215

12 T Sheppard and H McShane, Powder Metall., Vol 19 (No 3), 1976, p 121-125

13 T Sheppard and H McShane, Powder Metall., Vol 19 (No 3), 1976, p 126-133

14 T Sheppard, H.B McShane, M.A Zaidi, and G.H Tan, J Mech Work Technol., Vol 8, 1983, p 43-70

15 A Greasely and T Sheppard, Proc 2nd Int Conf Consolidation of Particulate Materials (Brighton), Aug

1975

16 A Kumar, P.C Jain, and M.L Menta, Powder Metall Int., Vol 19 (No 3), 1987, p 15-18

17 B Avitzur, Metal Forming: Processes and Analysis, McGraw-Hill, 1968, p 250-292

18 S.H Gelles, V Nerses, and J.M Siergiej, J Met., Vol 15 (No 11), 1963, p 843-848

19 P Loewenstein, E.F Jordan, and J.M Siergiej, "Extrusion of Unclad Beryllium," Internal Report, Nuclear Metals, Inc., 1961

20 A.P Green, Philos Mag., Vol 42, 1951, p 365

Powder Extrusion Practice

Extrusion of canned powder can be considered a special case of filled-billet extrusion Filled billet refers to an extrusion billet or workpiece that contains more than a single material For canned powder, there are at least two components, the powder and the can More typical filled-billet workpieces, discussed below, may have two or more different materials arranged in a geometric array within a container In the extrusion of a filled billet, it is important that the materials of the various components are metallurgically compatible A careful review of their mechanical and physical properties will avoid serious pitfalls, such as:

• The formation of eutectic phases

• Gross thermomechanical mismatch at extrusion temperature

• Excessive interdiffusion of can and workpiece material

• Extreme difficulty in removing the can material from the extruded product

These considerations are discussed in more detail by Roberts (Ref 5)

In the treatment of powder-filled billets, a further concern is the relative deformation of the canister and the powder contents during the upset of the billet The can may buckle in the press container if the resistance of the contents of the container to upsetting is low in comparison with the buckling resistance of the can The folds that result from buckling will carry through into the extruded rod as a defect This is shown schematically in Fig 11 and is discussed by

Trang 32

Loewenstein et al (Ref 3) The density of the powder and the wall thickness of the can must be balanced to avoid buckling One solution to this problem is to use a ram that penetrates the canister, densifying the powder without axially loading the canister wall This is shown in Fig 12 Alternatively, the powder billet may be precompacted as in Fig 3 or, more suitable for spherical powders, it may be jolted or vibrated during filling to increase the density of powder packing

Fig 11 Folding of metal can with loosely packed powder

Fig 12 Penetrator technique to avoid folding of can

Because densification is largely completed during the upsetting stage, extrusion of the canned powder can be thought of

as coreduction of two materials having a core-sheath relationship The relative resistance to deformation of the two materials must be balanced in order to avoid severe instabilities in plastic flow and possible fracture of either the sheath (can material) or core (densified powder workpiece) Avitzur has presented an analysis used to predict conditions that cause fracture in either the cladding or the core based on the following parameters (Ref 17, 21):

• Overall reduction ratio

• Extrusion die angle

• Volume fraction of each material

• Relative flow stress of each material friction

Roberts and Roberts (Ref 22) cite examples and conditions for successful coreduction of metals by hot extrusion and cold wire drawing

The extrusion constant K defined in Eq 1 can be used to design billets for either powder extrusion or extrusions of various

arrays of solids within a can The force required to coextrude a billet containing more than one material can be calculated

by a rule of mixtures approach:

Trang 33

F = (K1A1 + K2A2 + KiAi) ln R (Eq 5)

where the various K and A terms are the respective extrusion constants and the cross-sectional areas of the components that sum to A, the cross-sectional area of the press container A general guideline is to balance the required force components by selection of container material (K term) and relative area fractions of workpiece and container (A terms)

for the required reduction ratio

While precise matching of flow stress values is not practical, it serves as a simplified guide Gross mismatch of flow stress values usually results in unstable metal flow manifested by large variations in component sections along the extruded rod In extreme cases, discontinuities may occur Figure 13 shows the results of uneven reduction obtained from the coextrusion of seven cast rods of a cobalt alloy (Coast Metals alloys No 64, Co-30Cr-20W-1V-5Ni-1C) set in a drilled billet of low-carbon steel The billet geometry is shown in Fig 14 The reduction ratio and temperature used for this trial were 12 to 1 and 1065 °C The cobalt alloy rods were exposed by etching away the steel matrix in hot dilute nitric acid It is interesting to note that while this attempt to coextrude cast rods in a mild steel matrix was unsuccessful, much better results were obtained when the same cobalt alloy was introduced as -80 mesh (<175 m) powder

Fig 13 Attempt to coreduce cast cobalt alloy rods in low-carbon steel matrix See Fig 14

Fig 14 Low-carbon steel drilled billet with seven cobalt alloy rods introduced (see Fig 13) (a) Longitudinal

section B-B' (b) Cross section A-A'

In certain instances, trials have been conducted successfully that violate the simple rule of matching flow stresses An example involving a thin cladding of 10 carat gold alloy over a thin-walled tube of Inconel 600 with a central core of nickel rod and all contained within a steel can, extruded successfully with proportionate and uniform component sections

in the extruded product (Ref 5) Had the components been presented in the same amounts but with a different geometry of distribution, such as in the form of parallel rods of these materials, much less uniform coreduction of section could be expected The exact influence of component distribution is not well established, and a considerable amount of art is required for proper billet design

Trang 34

17 B Avitzur, Metal Forming: Processes and Analysis, McGraw-Hill, 1968, p 250-292

21 B Avitzur, Wire J., Vol 3, Aug 1970, p 42-49

22 J.A Roberts and P.R Roberts, U.S Patent 3,698,863, 1972

Materials Processed by Powder Extrusion

The early development of powder extrusion processes concentrated on the search for materials with improved temperature performance and the development of methods for encapsulating and handling toxic materials The use of an outer skin or container and a particulate workpiece also stimulated metalworkers to examine powder extrusion as a means for forming structural components from difficult-to-work material These basic uses of powder extrusion still drive the industry Examples are given below to illustrate the advantages of coupling powder metallurgy with a large deformation process such as extrusion for the production of unique materials

high-The most critical applications of extruded powders today involve extrusion of canned material high-The primary applications include tooling and aerospace markets, where the materials include tool steels, superalloys, titanium, copper, and aluminum Common to the products produced from these materials are the requirements for structural integrity and high performance These attributes are achieved only by proper powder processing and controlled extrusion practices

It is well established that shear deformation in combination with pressure fractures oxides and other films on particle surfaces and produces sound bonds between particles For this reason, extrusion of aluminum powders, which have substantial oxide contents, is common practice On a more subtle scale, extrusion of "clean" powders, for example, with oxygen levels below 100 ppm, relies on the same mechanism to produce particle bonds of high integrity Here, the environment within the sealed container is critical to the success of the process

Thus far, little mention has been made of the mechanical properties or microstructure of the extruded products because the process may be used to form a wide variety of materials for many applications A discussion of the versatility of the extrusion process as it applies to powder follows

Dispersion-Strengthened Materials. In the ongoing search for materials with improved properties at elevated temperature, sintered aggregates of metals and nonmetals were investigated This development started at the laboratories

of Aluminium Industrie AG in Switzerland in 1946 as an accidental discovery when experiments to make extruded Al-C wires from powders showed much higher strengths than expected The oxide skin that surrounds each aluminum powder particle inhibits sintering, and the result is a weak, brittle sintered product However, this skin is easily broken to expose clean metal surfaces that bond metallurgically during extrusion deformation to produce sound products Furthermore, extrusion tends to fragment and redistribute the oxide

Subsequently, Irmann (Ref 23) reported details of a commercially manufactured material that came to be known as sintered aluminum product (SAP) Irmann's development of SAP encouraged further research in the 1950s and 1960s A good account of the processing details and enhanced mechanical properties developed in the original SAP materials is found in Ref 24 Metallic systems with nonmetallic dispersion phases studied intensively included copper, cobalt, iron, nickel, silver, gold, platinum, and lead, using dispersions of oxides, carbides, and nitrides Zwilsky and Grant (Ref 25) mixed electrolytic copper with fine alumina; pressed, sintered, and extruded rods were swaged; and it was found that recrystallization was affected significantly A 10% addition of a 0.3 m alumina to -325 mesh (<45 m) copper powder

so processed was found to suppress recrystallization completely and retain useful hardness up to 1000 °C Cremens and Grant (Ref 26) mixed very fine alumina with fine carbonyl nickel Pressed, sintered, and extruded rod stock showed a sevenfold increase over pure nickel for stress rupture at 100 h In all of these experiments, extrusion was the primary densification process Ansell (Ref 27) suggested a mechanism whereby the supply of dislocation sources was diminished

by extrusion to explain the delay in the onset of creep and the often erratic creep behavior in these early strengthened materials He further suggested that extrusion or rolling may be essential manufacturing steps for these products

Trang 35

oxide-dispersion-Oxides of aluminum, magnesium, thorium, and the rare earths, such as yttrium, are generally preferred as dispersion phases because they are hard and stable at high temperatures Nickel strengthened with a dispersion of thoria is an early example of these products These early materials have been superseded by more well developed and precisely controlled dispersion-strengthened materials in recent years

The main methods of producing current dispersion-strengthened metals are mechanical alloying and chemical reaction, for example, internal oxidation (Ref 28, 29, 30, 31, 32, 33, 34, 35) Both of these techniques are based on powder metallurgy as a means of producing a uniform distribution of fine dispersoids, and both involve extrusion as the primary consolidation method Examples of these materials are presented below

Commercial grades of oxide-reinforced copper are available (C15715 with 0.3 wt% Al2O3 and C15760 with 1.1 wt%

Al2O3) in wire and rod form produced from extruded bar (Ref 32) Primary applications include spot welding tips that require high electrical conductivity and resistance to deformation at elevated temperature The oxide in this P/M product

is produced by internal oxidation, and it is redistributed during the large area reduction of the extrusion process At high temperatures, the material extrudes easily, and the extruded bar may be drawn to large reductions in section without intermediate annealing steps Consequently, the product is fine grained with high strength and yet maintains electrical conductivity close to that of pure copper The key to achieving a high elevated-temperature strength is the fine size (3 to

12 nm) and uniform spacing (50 to 100 nm) of the alumina dispersoid The combination of powder metallurgy and extrusion provides these metallurgical features

In the original aluminum SAP alloys, the oxide content was typically 12 to 15 wt%; high fractions of dispersoid were required because of the relatively coarse and uneven distribution Commercial dispersion-strengthened aluminum alloys currently available contain a significantly lower dispersoid content, that is, <5 wt%, but these alloys are much more effective than SAP alloys because of the fine size and the uniform distribution of the dispersoid (Ref 29, 30, 31) IN-9021 (Al-4.0Cu-1.5Mg-1.2C-0.75O) and IN-9051 (Al-4.0Mg-0.7C-0.60O) mechanically alloyed aluminum alloys contain oxide and carbide dispersions of less than 5 wt% Typical processing involves hot consolidation to densify the powder, followed by hot extrusion Extruded forms may be rolled, swaged, or drawn

Nickel alloy Inconel MA754 (Ni-20Cr-1Fe-0.3Al-0.5Ti-0.6Y2O3) is used for gas turbine vanes and is required to have a strong texture to provide good thermal fatigue and creep resistance (Ref 28, 33) Mechanically alloyed powder, attrited in

a high-energy ball mill, is thermomechanically processed via extrusion, hot rolling, and zone annealing to provide a strong <100> grain-axis alignment parallel to the working direction in order to impart the desired properties The directional working of extrusion and subsequent rolling or plane strain forging produces the microstructure needed to form a large, elongated grain structure during zone annealing to provide directional recrystallization

Dispersion strengthening of silver by particulate cadmium oxide was developed to produce electrical contacts having improved resistance to arc erosion, low contact resistance, and antiwelding characteristics (Ref 34) While the CdO content may be high in comparison with the dispersoid content in copper (e.g., up to 25% CdO in Ag versus 1% Al2O3 in copper), the thermal and electrical conductivities of these materials are excellent It is common practice to make rod stock

in this alloy by pressing, sintering, and extruding coprecipitated or preoxidized powders or powder blends

Platinum and its alloys, used in special furnace windings, heater tapes, and structural shapes that are required to operate close to their melting points, are fabricated from powder blends with thoria, yttria, or zirconia dispersions (Ref 35)

A dispersion-strengthened titanium powder produced by the plasma rotating electrode process (PREP) was consolidated

by extrusion (Ref 36) The chemistry of this alloy was a modification of Ti-6246 with an erbium addition 4Zr-2Mo-0.1Si-2Er) The rare earth was internally oxidized to Er2O3 in a fine dispersion of particles where the maximum size observed was 50 nm The principal intent of this exercise was to obtain fine dispersoids in a rapidly solidified powder material It is interesting to note that powder consolidation by extrusion did not compromise the useful metallurgical characteristics developed by rapid solidification processing of this material

(Ti-6Al-2Sn-These are some of the many examples of materials that have had their properties significantly extended by dispersions, and extrusion is most often used to consolidate the precursor powder blends into bar form that may then undergo further working Extrusion serves to densify the product and impart a uniform longitudinal and transverse distribution of the dispersed phase together with the development of preferred orientation in the metal matrix that assists further deformation

by wire drawing or other methods for the reduction of cross section

Trang 36

Superalloys. Extrusion is commonly used to consolidate superalloy powders into billet stock for subsequent forging operations (Ref 37, 38) Typical processing includes containerization of gas-atomized powder, hot compaction by forging

to around 94% of theoretical density, and extrusion to full density at a reduction ratio of 5-7 to 1 After can removal and inspection, the extruded bar is sectioned into billets for forging Extrusion is relied on to produce strong metallurgical bonds between powder particles by virtue of shear deformation and high pressure across particle interfaces In addition, extrusion produces a fine grain size that facilitates forging This is especially important in isothermal forging processes

Superalloy powders are susceptible to carbide formation on particle surfaces because of the presence of strong forming elements In high-carbon superalloys, such as René 41, carbides are relied on to provide grain size control and to enhance mechanical properties (Ref 38) In such cases, the extrusion temperatures must be controlled to avoid gross carbide formation before extrusion High extrusion preheat temperatures result in excessive carbide formation along prior particle boundaries in the extruded form that reduces transverse strength and ductility to unacceptable levels Furthermore, grain growth is limited to the powder particle size during high-temperature solution annealing, which results

carbide-in low elevated-temperature properties The use of a low extrusion temperature reduces the amount of particle-boundary carbide formation significantly Subsequent solution annealing temperatures can then be controlled to produce either a fine-grained microstructure with more uniform properties or a large equiaxed microstructure with good elevated-temperature properties

Unintentionally introduced organics are often the source of undesirable prior particle boundary decorations deformation consolidation processes, such as hot isostatic pressing, do not deform particles sufficiently to rupture surface films and present virgin surfaces for bonding Extrusion, by virtue of its large deformation, effectively removes the deleterious effect on performance of such contaminants when potentially harmful particle surface films are fragmented and isolated within the product

Low-In a novel application of superalloy powder filled billet extrusion, a method has been developed that relies on a solid configuration for fabricating thin-weld filler wire that is used for critical repair of turbine engine blades and vanes Weld wire has been made in the cobalt and nickel alloys listed in Table 2; adjustment of component dimensions and process parameters were derived to provide wire at various diameters While the technology is proprietary, the basic process is well known Arrays of tubes filled with clean spherical powder made by PREP are packed within a larger tube that is capped, evacuated, and sealed Because the billet contains both solid and powders, billet design must compensate for the differential in their packing densities and the effect this will have while the billet undergoes upset and the powder finally comes to full density PREP powder is preferred because it produces slag-free welds due to virtual absence of nonmetallic contaminants In addition, PREP spherical powders pack reproducibly to a density of 65% because particles are free of satellites The number of tubes processed and the high aspect ratio preclude ramming or step compaction for powder fill The billet is extruded to a diameter calculated to contain the desired wire diameter within the resultant rod Weld wires are ultimately released from the rod by immersion in a reagent bath that attacks the container matrix but is inert to the weld wire The cross section of a weld wire extrusion billet is shown schematically in Fig 15 After extrusion, the rod section contains an array of evenly spaced wires consolidated to full density The wire sections are not perfectly circular, but the total variation in cross-sectional area from wire to wire and along the length of any wire does not exceed 5% An example of an extrusion section and some short samples of released wires are shown in Fig 16

powder-Table 2 Weld wires fabricated by filled billet technique

max

0.1 max

Trang 37

Fig 15 Cross section of extrusion billet to fabricate superalloy weld wires

Fig 16 Extrusion cross section and some released weld wires of cobalt alloy (Coast Metals No

64-Co-30Cr-20W-5Ni-1V-1C) These wires were made from -80 mesh (<175 m) powder, cf performance of cast material

in Fig 13)

Beryllium. Extrusion of beryllium serves as an excellent example of processing a toxic, difficult-to-work metal into useful structural shapes Beaver and Wikle (Ref 39) have reported comprehensive mechanical properties for beryllium This metal has a hexagonal close-packed crystal structure and shows the development of a considerable degree of anisotropy when hot-pressed compacts are hot worked by extrusion Most notably, the ductility is enhanced in the working direction Gelles et al (Ref 18) described the texture developed by extrusion, which imparts the increased elongation properties in the extruded direction Basal planes tend to lie parallel to the extrusion axis with the <1010> crystallographic direction aligned Fracture occurs by cleavage on the (1120) planes

The much-improved properties obtained by extrusion have resulted in the application of beryllium tubing in the construction of communication satellites The combination of low density (1.85 g/cm3) with a high modulus of elasticity and an elongation in excess of 10% in the longitudinal (extrusion) direction make this an ideal structural material for this application

Loewenstein et al (Ref 19) have concluded that while it is possible to extrude beryllium uncanned using a glass lubricant, after the method of Ugine-Séjournet (Ref 40, 41), cracking will occur in any section more complex than a simple round rod, unless considerable precautions are taken, especially in die design In view of the toxicity of the metal and its proclivity to oxidize, any advantage that might be realized in dimensional tolerance control via bare extrusion would seem

to be outweighed by the hazards inherent to this practice

Trang 38

Normal canning for beryllium powder billet construction is low-carbon steel, and this is readily removed by nitric acid when the beryllium core is to be released The use of pressed powder rather than cast material as the precursor beryllium charge produces much finer grain size in the extruded product (Ref 4) The grain refinement attained contributes to the improved ductility in the extruded product A detailed review of beryllium extrusion practices is given by Loewenstein (Ref 42)

Ferrous Alloys. Although powder extrusion is intrinsically an expensive process, benefits in improved homogeneity and a more direct route from melt to a wrought, consolidated structure are realized for a number of materials that, until recently, have been wedded to conventional practice involving casting and extensive hot working An example is the manufacture of fully dense stainless steel tubes for critical service, such as in the production of urea carbamate Two stainless steel compositions, referred to as 724L and 724LN, are manufactured by Nyby Uddeholm AB, Torshalla, Sweden, and the compositions are given in Table 3 (Ref 43) Inert gas atomized powder is canned in mild steel, with a packing density of 70% of theoretical being achieved The powder billets are cold isostatically pressed to densify the powder further to 85 to 90% of theoretical density These are then extruded at 1200 °C using a glass lubricant to reduce friction Tubing is fully densified for extrusion ratios at or above 4 to 1 The use of powder extrusion to produce tubing and pipe seems to be a logical extension of the technology developed to produce rolled sheet directly from stainless steel and other powders

Table 3 Stainless steel compositions (wt%) extruded into tube by Nyby Uddeholm AB

Composite Materials. Powder metallurgy coupled with extrusion offers a unique method of fabrication of both macro- and microcomposite materials Erich (Ref 46) reviewed P/M applications for the production of metal-matrix composites Advantages of a P/M approach in comparison with fusion metallurgy methods include low processing temperatures, unique materials, use of particulates or whiskers, and a wider capability for in situ formation of strengthening phases Canned extrusions have been developed to fabricate powder mixtures and multicomponent systems that include arrays of both solid sections and powders The reinforcing agent may be formed in situ by metallurgical reaction, or it may be introduced as a separate component during powder blending Extrusion serves to align the reinforcing agent as well as consolidate the product

Intimate mixtures of metal powders sealed within an evacuated canister have been extruded to make fibers having high aspect ratios Rebundling of the primary extrusion into a second can, such as the method of Klein (Ref 47), is capable of producing filaments that are extremely fine Depending on the application, these may be released by differential chemical attack or the composite structure formed may be the desired product

Composite rods extruded from copper-niobium and copper-tin bronze-niobium powder mixtures are described by Foner (Ref 48, 49) These rods were drawn to wires using large section reductions, and the niobium particles were coreduced proportionally to small diameters and long lengths Appropriate heat treatment of the bronze matrix wires resulted in the formation of the intermetallic compound Nb3Sn having a tungsten (A15) structure and excellent hard superconducting

properties

Although the superconducting filaments are long, they are discontinuous and therefore might not be expected to impart superconducting properties to the wire However, extensive reduction in cross section will provide small filament

Trang 39

diameters with concomitantly finer interfilament spacing; a careful selection of particle sizes and section reduction schedules can provide interfilament spacing that is sufficiently small to permit electron pair tunneling as described in the theory of Bardeen et al (Ref 50) The highly desirable fine diameters of the filaments together with this tunneling effect combine to produce excellent type II superconducting properties in the wire when it has received suitable processing In addition to the fabrication of superconductors, this type of structure has other interesting applications Foner (Ref 49) cites the possibility of making magnetic wire materials with this method, and indeed Levi (Ref 51) describes the formation of such wire structure involving filaments of iron in a copper matrix made by successive wire drawing and bundling steps Although Levi uses neither powders or extrusion, the principle for successful coreduction of mechanically compatible materials is demonstrated, and structures offering a similar degree of reinforcement are feasible by the method of Klein (Ref 47)

Powder mixtures and powders mechanically mixed with chopped fibers can be extruded to produce fiber-reinforced structures If all the components are metallic and possess similar flow strengths at the working temperature, they will elongate similarly as described by Roberts et al (Ref 5, 22) and Klein (Ref 47) Ultimately, long fibers of one species in a matrix of another may provide a potent strengthening mechanism Where the fibers and matrix have dissimilar elastic moduli, with that of the fiber being greater than that of the matrix, a complex stress distribution is developed during axial loading where the shear stress developed at the fiber/matrix interface and the axial tensile stress in the fiber may be represented graphically as shown in Fig 17 Essentially, this figure shows that the fiber carries a tensile load produced by the shear stress developed along the fiber matrix interface High allowable interface shear stresses permit the fiber to carry high axial tensile stresses Shear stresses above some critical value at the fiber ends cause the matrix to deform plastically

To obtain the best use of the high-strength fiber, the plastic zone in the matrix should not extend from the fiber ends to its

midpoint before the fiber is overstressed A critical fiber length Lc is given by:

where f is the fracture stress of a fiber of diameter d in a matrix having a shear yield stress o When the fiber length is

greater than Lc, composite failure occurs by fiber fracture and the system has developed its full strengthening potential

The critical aspect ratio Lc/d for fibers determines the minimum coreduction of section ratio; for all extrusion processing this will be the minimum value for R, the extrusion reduction ratio

Fig 17 Stress distribution along embedded fiber caused by uniaxial loading

Erich (Ref 46) cites an example of in situ composite fabrication by extrusion where tungsten powder was blended into a nickel powder matrix, compacted, and sintered This blend can be sintered with a transient liquid phase technique The sintered billet is then extruded and may be further worked by rolling or drawing to produce a composite with a microstructure of highly elongated tungsten grains in a nickel matrix The resulting product is reported to equal the strength of a directionally solidified material of similar chemistry

Trang 40

18 S.H Gelles, V Nerses, and J.M Siergiej, J Met., Vol 15 (No 11), 1963, p 843-848

19 P Loewenstein, E.F Jordan, and J.M Siergiej, "Extrusion of Unclad Beryllium," Internal Report, Nuclear Metals, Inc., 1961

22 J.A Roberts and P.R Roberts, U.S Patent 3,698,863, 1972

23 R Irmann, Tech Rund., Vol 41 (No 36), 1949, p 19

24 Symposium on Powder Metallurgy, The Iron and Steel Institute, London, 1954, 1956

25 K.M Zwilsky and N.J Grant, J Met., Vol 9 (No 10), 1957, p 1197-1201

26 W.S Cremens and N.J Grant, Proc ASTM, Vol 58, 1958, p 714-730

27 G.S Ansell, Trans AIME, Vol 215, 1959, p 249-250

28 J.S Benjamin and T.E Volin, Metall Trans., Vol 5 (No 8), 1974, p 1929-1934

29 J.S Benjamin and M.J Bomford, Metall Trans., Vol 8A, 1977, p 1301-1305

30 G Jangg, F Kutner, and G Korb, Powder Metall Int., Vol 9 (No 1), 1977, p 24-26

31 K.H Kramer, Powder Metall Int., Vol 9 (No 3), 1977, p 105-112

32 A.V Nadkarni and J.E Synk, Powder Metallurgy, Vol 7, ASM Handbook, American Society for Metals,

1984, p 700-716

33 P.S Gilman and J.S Benjamin, Powder Metallurgy, Vol 7, ASM Handbook, American Society for Metals,

1984, p 722-727

34 J.E Synk, Powder Metallurgy, Vol 7, ASM Handbook, American Society for Metals, 1984, p 716-720

35 Anon, Powder Metallurgy, Vol 7, ASM Handbook, American Society for Metals, 1984, p 720-722

36 F.H Froes and R.G Rowe, Rapidly Solidified Alloys and Their Mechanical and Magnetic Properties, B.C Giessen et al., Ed., Symp Proc., Vol 58, Materials Reserarch Society, 1986, p 309-334

37 B.L Ferguson, Powder Metallurgy, Vol 7, ASM Handbook, American Society for Metals, 1984, p 646-656

38 G Friedman, Int J Powder Metall Powder Technol., Vol 16 (No 1), 1980, p 29-35

39 W.W Beaver and K.G Wikle, Trans AIME,Vol 200, 1954, p 550-573

40 British patents, 607,285; 661,555; and 663,357

41 J Séjournet, Iron Coal Trades Rev., Vol 165, 1952, p 963

42 B Loewenstein, Beryllium Science and Technology, D Webster et al., Ed., Vol 2, Plenum, 1979, p 67-82

43 C Aslund, Prog Powder Metall., Vol 39, 1983, p 543-558

44 N Kawai and H Takigawa, Met Powder Rep., Vol 37 (No 5), 1982, p 237-240

45 R.S Carbonara, State of the Art Review of Rapid Solidification Technology (RST), Report MCIC 81-45, Metals and Ceramics Information Center, 1982, p 36

46 D.L Erich, Int J Powder Metall., Vol 23 (No 1), 1987, p 45-54

47 J.L Klein, U.S Patent 3,413,707, 1968

48 S Foner, Prog Powder Metall., Vol 38, 1982, p 107-114

49 S Foner, Met Powder Rep., Vol 39, 1984, p 583

50 J Bardeen, L.N Cooper, and J.R Schrieffer, Phys Rev., Vol 108, 1957, p 1175

51 F Levi, U.S Patent 3,029,496, 1962

Innovative Applications

Định dạng
Số trang	160
Dung lượng	3,45 MB