Seetharaman S. Fundamentals of metallurgy

Подождите немного. Документ загружается.

There are two types of plasma, thermal plasma and cold plasma, depending

on the respective temperature of the electrons (T

) and the heavy particles

(neutral particles and ions (T

)). In thermal plasmas (pressure > 10 kPa), the

temperatures of the gas and the electrons are comparable and are of many

thousands of degrees. Starting materia ls are typically ato mised at these

temperatures and the powder synthesis occurs during condensation outside the

plasma when the gasses cool.

In contrast, in a low-pressure (p < 200 Pa) low-

temperature plasma (cold plasma) the electron temperature is much higher than

the gas or ion temperature, which is close to room temperature because of the

poor collisional coupling betwee n electrons and heavy particles at reduced

pressures.

As the growth of particles may be completely determined by

chemical kinetic factors, thermal and low-temperature plasmas may produce

materials with different structures and properties.

Plasma methods are very popular as they enable the preparation of two-

component compounds as well as multicomponent powders with a high purity.

Besides ceramic powders (SiC, SiN, BN), metallic nanoparticles are produc ed.

Nano-powders are used for their light emission properties (Si) or magnetic

recording tapes. Nanostructured thin fil ms, consisting of nanoparticles

embedded in a metal matrix have also been produced.

12.3 Forming processes towards near-net shape

Near-net-shape components can be obtained by a conventional two-step process

consisting in first giving the shape of the component (compaction) followed by

its consolidation (si ntering). Other processes allow the shaping and the

consolidation of the component in one step (hot isostatic pressing, hot extrusion,

hot forging, etc.). Because of its economical importance infiltration is also

shortly described in this part, as well as the fast developing rapid prototyping

processes.

12.3.1 Conventional route

Compaction

Compaction is a widely used method to form semi-finished or near-net-shape

components from a powder. The quality of compaction (density gradients,

micro-cracks) has a great influence on distortion of the product shape after

sintering step as well as on the properties (defect-free components) of the final

product.

Powder compacts can be obtained by unia xial pressing, isostatic pressing,

metal injection moulding (MIM) or by less used processes such as powder

rolling, extrusion and dynamic or explosive compaction.

486 Fundamentals of metallurgy

Uniaxial pressing

Uniaxial pressing (dry or die pressing) is the most common method of

compaction to form PM components. This low-cost process is adapted to high-

volume (up to few hundreds of parts per minute) production of `relativel y

simple' geometry powder compacts.

It consists in compacting a dry powder (i.e. < 2 wt% water) in a die at a

pressure ranging from 20 to 700 MPa by operating one or more rigid punches

(higher numbers of punches are needed with increasing complexity of the

component). A sufficiently high pressure is required to guarantee a sufficient

strength of the green compact for subsequent handling and processing.

Compaction consists in the following three steps: filling the die with the

powder, compacting the powder and ejecting the powder compact from the die.

Die filling and compaction control the uniformity of the green density in the

powder compact, which is a crucial parameter. Green density gradients have to

be minimi sed in order to reduce differential shrinkage during sintering and shape

distortion, as well as to avoid induced defects that limit the properties and

reliability of the part. Ejection of the compact is also a critical step because

macroscopic defects can be created.

Uniform and high packing density of a powder during die filling favours high

green density with low green density gradients after compaction. This implies

the use of good-flowing powders, which are also necessary to guarantee

reproducible die filling. High-flowability powders have a relatively large

particle size distribution lying between 40 and 400 m and are usually spherical

or equiaxial. Powders with a low flowability (fine (1±10 m) irregular powders)

have to be granulated before dry pressing (e.g. by spray drying) with 1 to 5 wt%

organic additives such as binders, plasticisers and lubricants, which also

improve the handling of the compact and/or the compactibility of the powder.

Organic binders (wax, polyethylene glycol) allow the increase the green strength

of the compact. Plasticisers (water, ethylene glycol) are used in combination

with binders to improve the deformability of the powder during compaction.

Lubricants (wax, magnesium stearate, stearic acid) reduce the interfacial

frictional forces between individual particles favourable to powder compaction

and/or between particles and die surfaces. They also reduc e the required ejection

pressure of the compact, thus avoiding macro-defect formation. Those additives

will be eliminated by thermal decomposition before sintering or in a special

zone of a continuous sintering furnace.

During powder compaction, the applied pressure induces particle

rearrangement, deformation, in som e cases fracturing, and finally consolidation

of the particulate assembly. Green density and compact strength increase with

increasing compaction pressure.

Two scales of green density gradients can

occur during powder compaction: (i) macros copic density gradient induced by

non-uniform die filling and/or pressure gradients during compaction and (ii)

Understanding and improving powder metallurgical processes 487

microscopic density gradient due to packing defects, hollow particles (granules),

and/or insufficient particle (granule) deformation during compaction. Non-

uniform pressure/stress gradient is due to the non-uniform pressure/stress

transmission at particle±partic le and particle±die (die wall friction) contacts.

Green density gradients are minimised by using lubricants, smooth surf ace die,

low compaction ratio, double-action or floating die pressing instead of single-

action pressing and by adapting the numbers of punches to the number of

different thickness areas in the part.

During ejection, the compact integrity is favoured by a sufficiently high green

strength to withstand the applied forces on the compact required to eject it from

the die, a low ejection pressure, favoured by the use of lubricant to reduce die

wall friction and a low elastic springback (expansion of the powder compact

after ejection from the forming die). The negative effect of a differential

springback (the axial springback being higher than the radial one) can be

reduced by maintaining a small axial pressure on the compact during ejection

(punch hold-down ejection or withdrawal of the die).

Uniaxial pressing can be performed with a single-action, double-action or

floating die press (Fig. 12.4).

A single-action press uses only one moving punch. As it induces high green

density gradients in the axial directions, it may be used for parts with a very low

height to diameter aspect ratio and simple shape. A double action press has at

least two independently moving punches that induce a more uniform compaction

12.4 Uniaxial pressing: (a) single action, (b) double action, (c) floating die and

double action multiple punches.

488 Fundamentals of metallurgy

especially for simple large and thick parts. In pressing of multi-level compacts,

the several punches have to be operated independently so that the different areas

in the die are equally compacted. In a floating die press, a similar effect can be

obtained, the movement (floating) of the die corresponding to an upwards

movement of the lower punch.

In more sophisticated floating die press, the

punches and the die can move independently in order to reduce the green density

gradients. This allows the production of complex parts with tapers, holes and

multiple steps. If higher working speeds and precise control of the movements

are required, the die is directly moved by the press.

Isostatic pressing

Isostatic pressing (cold isostatic pressing) consists in compacting a powder in an

elastomeric contain er submersed in a fluid at a pressure of 20 to 400 MPa. Cold

isostatic pressing allows the production of simple-shaped small or large powder

compacts (up to 2000 kg) with a uniform green density even for large height/

diameter ratio part (impossible by uniaxial pressing), but with the sacrifice in

pressing speed and dimensional control, requiring subsequent machining in the

green compact. Cold isostatic pressing is used for powders that are difficult to

press such as hard metals.

Metal injection moulding

Metal injection moulding (MIM) is adapted to the production of small complex

near-net-shape compounds (wall thickness down to 0.3 mm) with very good

tolerances (0.3 to 0.5%), especially for medium (thousands of parts/year) to high

volume (millions of parts/year) production.

Because of the high raw-material

costs, MIM is usually limited to the production of parts lighter than 100 g.

This process consists in filling a die cavity with a viscous mixture of powder

and binder at around 130 to 200ëC under a pressure up to 150 MPa.

Metal

injection moulding parts are produced in four different steps: feedstock

formulation, moulding, debinding and sintering.

The feedstock is a homogeneous pelletised (granulation to a special shape)

mixture of fine spher ical metal powder with a mean size ranging from 5 to

15 m and 30 to 45 vol.% organic binder.

A binder system usually has three

components, a backbone that provides green strength and that will stay after

debinding, a filler phase that is easily extracted during initial stages of debinding

and surfactant to control feedstock rheology. During moulding, the feedstock is

introduced to the heated injection unit by, for example, a rotating screw, the

mixture is then injected in the die by axial movemen t of the screw and finally the

part is ejected from the die after quick cooling. The main parameters are

injection temperature, pressure and speed, as well as thermal conductivity and

viscosity of the feedstock and the mould temperature.

Understanding and improving powder metallurgical processes 489

The debinding can be done by thermal decomposition, solvent extraction or a

combination (also called catalytic decomposition). A typical binder removal rate

is 2 to 3 mm wall thickness per hour. Finally, the part undergoes a densification

up to 95% to 99% of the theoretical density as well as a shrinkage of 14% to

20% during the sintering step (see below).

The MIM process is applied to produce parts in stainless steels (304L, 316L),

tool steels (M2), soft magnetic alloys (Fe-50 %Ni, Fe-3 %Si, 430L), alloys for

glass-to-metal sealing applications (kovar), cobalt-, nickel- and titanium-based

alloys used for wear medical, automotive and aerospace applications.

Powder rolling ± extrusion ± dynamic and explosive compaction

Powder rolling (roll compacting) consists in compacting a powder continuously

passing between two turning rolls.

A binder is usually added to the powder to

favour densi fication. For a given thickness of the strip (defined by the distance

between the two rolls), the density of the compacted strip can be increased by

the increasing the diameter of the roll and reducing the rolling speed (maximum

speed lying about 0.5 m/s). The main advantage of powder rolling compared to

conventional casting and rolling process is a low amount of rolling passes

necessary to produce a thin strip. The main disadvantages are a high powder

price and a low production rate.

Extrusion consists in forcing a viscous mixture of powder and binder through

a die. The obtained, shaped product is sintered with a slow heating rate in order

to remove the binder.

During dynamic and explosive compaction (high-energy rate compaction),

that is only applied at lab-scale, powders are compacted at very high velocities

(200 m/s) by the propagation of a high-pressure wave. One set-up uses the

conventional die compaction with an upper punch moving at high velocity

through the action of an explosive charge or compressed gas. Another one

consists in encapsulating a powder in a mild steel tube and subjecting this tube

to the action of sheet explosives taped to it.

The advant ages are a high green density, a higher green and sinter ed strength

and lower density gradient.

13,14

Green density of 99% of theoretical density has

been achieved for aluminium, stainless ste el, amorphous powders,

even for

tungsten compacts a relative density of 97.6% has been obtained.

Sintering

Sintering originally used to produce clay pots

is nowadays involved in the

fabrication of net-shape components in ceramics, cermets, metals and com-

posites. Th e application fields are automobile and aeronautic industries (valves,

bearings, aircraft wings weight), electrical and electronic industries (tungsten

wires, ultrasonic transducers), medical industries (dental or hip implants).

490 Fundamentals of metallurgy

During sintering porosity and the microstructure change irreversibly from

contacting particles to almost dense material. This induces improvement of

many properties such as strength, ductility, conductivity, magnetic permeability,

and corrosion resistance.

Sintering can be defined as a thermal treatment for bonding particles into a

coherent, predominantly solid structure via mass transport events that often

occur on the atomic scale.

The sintering induces the consolidation (increase of

strength) of a loose or compacted powder and is usually accompanied by

densification (shrinkage). During sintering the reduction of total surface energy

(usual driving force for mass flow during sintering) is due to the decrease of

surface area by formation of inter-particle bonds and the reduction of surface

curvature. The path of the atomic motion occurring during the mass flow in

response to the driving force is called the sintering mechanism.

For metal

powders, the mechanisms are usually diffusion processes with surface, grain

boundary or lattice paths. Sintering progr esses in different stages; for each stage

(i.e. for each driving force type corresponding to a particular particle-pore

geometry), different ways of mass flo w (i.e. sintering mechanism) are possible.

The knowledge of the relations between the different sintering parameters and

each sintering mechanism during the different sta ges allows modelling and thus

optimising the sintering parameters.

The main sintering parameters are:

· particle size (reduction of particle size increases the surface energy per unit

volume of the powder and so the driving force associated with sintering, thus

the sintering rate);

· particle size distribution (large difference in curvature of the grains, due to

grain size difference, will promote coalescence, i.e. the growth of the large

grain at the expense of the smaller one);

· temperature (exponent ial influence on sintering because it is involved in the

activation energy of the sintering mechanisms (e.g. diffusion processes));

· time (influences diffusion);

· green density (density gradients occurring during compaction will induce

differential shrinkage during sintering and maybe distortion of the part

because higher green density induces lower shrinkage);

· applied external pressure (to obtain fully dense material, HP or HIP

processes);

· amount of liquid, if any;

· sintering aids (favouring diffusion in the solid or liquid state);

· atmosphere.

A particular sintering atmosphere can be used to protect the metal powder from

oxidation (argon, vacuum), to remove the oxidati on layer on the powder

(reducing atmosphere such as hydrogen, carbon monoxide, dissociated ammonia

or natural gas), to control the carbon content of the powder, to remove the

Understanding and improving powder metallurgical processes 491

lubricants and binders introduced duri ng compaction or even to react with the

powder (form ation of nitride). Sintering can be classified as solid state, liquid

phase, reaction sintering and microwave sintering.

Solid state sintering

In solid state sintering the microstructure changes are divided into three different

stages. The first stage corresponds to the growth of the bond, called neck, between

two particles, independently of the growth of the neighbouring necks. The pores are

interconnected with an irregular shape. The intermediate stage occurs as the

merging necks shrink the pores to form interconnected pores with a more smooth

usually cylindrical shape. Most of the densification and change in properties occurs

in this intermediate stage. The final stage corresponds to pore closure, where the

pores become spherical and isolated. By definition the driving force changes for

each sintering stage. During the initial stage the driving force is the curvature

gradient between the particle and the neck, during the intermediate stage it is the

curvature around the cylindrical pore and during the final stage the curvature

around the spherical pore. For every sintering stage, the mass transport process

(sintering mechanism) can be described by a characteristic equation.

There are sintering mechanisms that induce shrinkage (i.e. densification) such

as volume diffusion, grain boundary diffusion, plastic flow and viscous flow (for

the amorphous solids) and some that do not such as surface diffusion,

evaporation±condensation and volum e diffusion from a surface source to a

surface sink. Different sintering mechanisms are involved at different moments

during the sintering process. For example, a finer particle size usually favours

sintering by surface or grain boundary diffusion compared to volume diffusion.

During the initial stage, the different sintering mechanisms can be diffusion

(surface-, volume- or grain boundary-diffusion), evaporation or disloca tion

motion. The sintering mechanism is usually described by the size of the growing

neck between the particles but shrinkage or relative change in surface area can

also be used.

The intermediate stage is characterised by densification usually coupled with

grain growth during the latter phase of the intermediate stage. The smaller

grains, having a higher curvatur e, are progressively incorporated to the

neighbour grain by grain boundary motion. This grain boundary motion induces

drag forces on the pores, which can move by volume, surface diffusion or

evaporation±condensation mec hanism across the pore. When the moving rate of

the grain boundary is too high (e.g. favoured by high temperature), the pores

cannot impinge the grain boundary any more and become isolated inside the

grain (beginning of the third stage). In this case, the rate of densification is much

smaller because volume diffusion is less fast than grain boundary diffusion.

Consequently, it is important to minimise pore±grain boundary separation by

careful temperature control, the incorporation of second-phase inclusion such as

492 Fundamentals of metallurgy

oxide particles into the microstructure to impinge the grain boundary, or the use

of narrow initial particle size distribution.

37,38

The usual parameter to follow the

sintering is the rate of densification. The two sintering mechanisms involved in

densification are grain boundary and volume diffusion. Surface diffusion or

evaporation±con densation mechanisms, inducing no shrinkage, are also

expected to be active in smoothing the pore structure and in pore migration

with grain boundaries during grain growth. Long sintering times (compared to

the first stage) are required to achieve significant property or density change s.

Temperature has a complex effect on the sintering because diffusion, grain

growth and pore motion are all thermally activated.

The third stage is characterised by the presence of isolated spherical pores. If

the closed pores are mobile enough to stay coupled to the grain boundary, then

continued shrinkage is expected. This is favoured by a homogeneous grain size,

which lowers the curvature of the grain boundary and so decreases their motion

rate. If not, after separation from the grain boundary, the pore must emit

vacancies that move by volume diffusion, which is a slow process, towards the

distant grain boundary. This leads to a drop of the densification rate. With

prolonged sintering, the larger pores grow at the expense of the small er ones

(that emit more vacancies in the grain because of higher curvature). This is

called pore coarsening or Ostwald ripening. In addition to pore coarsening, the

pore size can increase by coalescence, due to grain growth by grain boundary

motion dragging pores towards each other. If the pore has trapped gas, an

internal pressure is induced inside the pore, which limits densification. If this gas

is soluble in the matrix, the densification rate is controlled by the internal gas

pressure and not by the limit of solubility.

The usual parameter to follow the sintering during the final stage is the rate of

densification, but the rate of shrinkage, surface area change, or neck growth

could also be used. The rate of densification depends on the pore amount, pore

radius, volume diffusion, grain size distribution and stress effects (compressed

trapped gas workin g against pore shrinkage).

Liquid phase sintering (HSS)

Liquid phase sintering is involved when powders of different composition are

mixed. Usually the constituent, that remains solid during sintering, should have

a relatively high solubility in the formed liquid and inversely the solubility of the

liquid in the solid should be low to ensure that this liquid phase is not transient.

Common systems are: WC-Co, Fe-Cu, Cu-Sn, etc.

The main advantage of liquid phase sintering is the lower sintering time

required compared to solid state sintering. During heating, the mixture of

powders first undergoes solid state sintering, which can induce significant

densification, before the formation of the liquid at the sintering temperature.

After the liquid is formed, the sintering depends on the amount of liquid and is

Understanding and improving powder metallurgical processes 493

usually divided into three stages: rearrangement, solution±r eprecipitation, and

the third stage. If the amount of liquid is sufficient to fill all the interparticle

spaces, the theoretical density can be obtained during the rearrangement stage.

For lower liquid contents, the solid skeleton slows down the densification and the

contribution of the last two sintering stages becomes significant. In fact, less than

15 vol.% of liquid is usually used to avoid distortion of the part during sintering.

During the first stage (rearrangement) and in case of a wetting liquid, the

liquid spreads as soon as the liquid is formed between the solid particles under

the influence of capillary forces. The rate of densification controlled by viscous

flow is very high at the begi nning and then continuously slows down. As the

densification rate governed by rearrangement decreases, another mechanism

called dissolution±diffusion±reprecipitation, characteristic for the second stage

prevails. The solubility of a grain in its surrounding liquid increases with the

curvature of the grain, i.e. with a decrease of the grain size. The difference of

solubility as a function of grain size induces a concentration gradient of solute

species in the liquid, that diffuse from the small grains to the large ones, on

which the solute species precipitate (reprecipitation) when the solubility limit in

the liquid is reached. So during this stage, grain coarsening occurs (Ostwald

ripening). Simultaneously the elimination of the high-energy vapour interface is

ob tained by grain shape accommodation (fl attening) during solution±

reprecipitation events inducing densification by higher packing of the grains.

The third stage corresponds to the densification of a rigid solid skeleton with a

rate similar to the one obtained in solid state sintering.

Reaction sintering

Solid state reaction sintering of metal powder mixtures is a way to produce

alloys such as carbon steel, Fe-Ni, Fe-Mn, Fe-Si, Fe-Cr, Fe-Mo and Cu-Ni. The

principal phenomenon occurring during sintering is the solid state interdiffusion

between the different compounds. The driving force is the chemical potential

gradient due to concentration differences. This phenomenon superimposes the

metal powder self-diffusion caused by surface and interfacial tension forces,

occurring in solid state sintering of pure or pre-alloyed powders. Reaction

sintering is favoured by fine particle size (smalle r the diffusion distance) and

high temperature (higher coefficient of diffusion).

Reaction sintering can also occur during liquid phase sintering (Mo + 2Si !

MoSi

) or by reaction of the sintering atmosphere with the powder (3Si + 2N

Microwave sintering

Sintering of metal powders is a surprising recent development in microwave

applications because bulk metals reflect microwaves. However, compacted

494 Fundamentals of metallurgy

metal powders at room temperature will absorb microwaves and will be heated

very effectively and rapidly (above 100ëC/min), inducing sintering. Metals such

as iron, steel, copper, aluminium, tin, nickel, cobalt, tungsten have been sintered

to high density by microwaves. Cylinders, rods, gears, and other automotive

components with until now a maximum size of 10 cm have been produced in 30

to 90 min.

Contrary to the conventional heating where the transfer of thermal energy is

done by conduction to the inside of the part, microwave heating is a volumet ric

heating consisting in an instantaneous conversion of electromagnetic energy into

thermal energy. The mechanisms involved in microwave sintering are not

completely understood at the moment. However, the sample size and shape, the

distribution of microwave energy and the mag netic and electromagnetic field

radiation are important parameters.

12.3.2 Other routes

This part is mainly dedicated to full density sintering processes such as hot

pressing, hot isostatic pressing, hot extrusion, hot forging and field assisted

sintering. However, other techniques such as infiltration and rapid prototyping

techniques will also be presented.

Full density is required to improve product properties such as rupture and

fatigue strength, toughness, thermal or electrical conductivity. This can only be

achieved when stress and temperature are simultaneously applied during den-

sification to close the pores.

Because the powder usually has to be protected

from reaction with air, this makes the hot compaction processes complex and

costly. So these techniques are reserved for expensive materials with special

properties such as beryllium and magnesium alloys (finer grain size), super-

all oys (elimination of segregation), high speed steels and dispersion

strengthened alloys (homogeneous distribution of the second phase). The main

parameters are temperature, applied stress, strain rate and grain size.

`Near-net-shape' components, implying material saving, combined with

higher properties can make these processes competitive compared to the

conventional casting, forging, machining route.

Hot pressing

Hot pressing consists in applying pressure with a hot punch on the metal powder

placed in a heated die usually under a protective atmosphere.

The main

problem with hot pressing is to find a suitable die material, which has to

withstand the applied pressure without reaction with the metal powder. Although

the total amount of deformation of the compact is relatively limited compared to

hot extrusion or hot forging, complete densification is generally achieved.

Understanding and improving powder metallurgical processes 495