Seetharaman S. Fundamentals of metallurgy
Подождите немного. Документ загружается.
(established by equilibrium with the partial pressure in the gas). D is the
diffusion coefficient of the dissolved impurity element in the melt.
When a boundary layer is absent, a model has been developed (Danckwerts,
1951; Machlin, 1960; Darken and Gurry, 1952) that assumes that rigid body
elements of fluid move parallel to the melt±gas interface, and during their time
of contact with the interface, they get degas sed. The streamline flow of the
elements is assumed to be free of convective currents and within the elements,
semi-infinite diffusion is assumed to be valid. The rate of degassing, is then:
dC
dt
dM
dA
dA
dt
1
V
1:90
Here dA/dt is the rate at which gas±melt interface is created and V is the volume
of the melt. dM/dt is the change in solute per unit area of gas±metal interface,
and can be expressed (Darken and Gurry, 1952) as:
dM
dA
1:1284C
o
ÿ CDt
0
1=2
1:91
where t
0
is the time interval during which solute is lost from the fluid element.
Due to the dependence upon diffusion, vacuum degassing is a very slow and
expensive method and is therefore primarily used for applications where high
purity is needed.
Electro slag remelting
In this refining method, an impure ingot is used as an electrode that is immersed
into a molten, ion-conducting slag consisting of CaO, Al
2
O
3
and CaF
2
. A high
current is passed through the ingot, whereupon the ingot melts. During this
process, inclusions dissolve in the slag. The molten metal, now purified from
inclusions, drops through the slag, and resolidifies as an ingot in the bottom in a
water-cooled mold. The water-cooled mold allows for a skull to be formed and
thus enables the refining of reactive molten alloys that cannot be easily contained.
Zone-refining
Zone refining is used to produce primarily materials for electronic components
such as Si or Ge. The principle is to utilize the fact that most impurities have a
higher solubility in the melt than the solid metal (Fig. 1.5) and thus partition into
the melt ahead of the solidification front during solidification. During zone
refining, an impure bar is passed slowly through a hot zone that melt s a small
section of the bar as shown in Fig. 1.7. As the melt zone is moved along the bar,
impurity successively builds up in the melt as shown in Fig. 1.8.
A mode l developed by Pfann (1958) describes the concentration in a bar, with
initially a uniform impurity concentration of C
o
during a single pass.
26 Fundamentals of metallurgy
C
C
o
1 ÿ 1 ÿ ke
ÿkx=l
1:92
Here, C is the impurity concentration, k is the partition coefficient of the
impurity between the solid and liquid metal, x is the length solidified and l is the
length of the molten zone. Several passes are carried out to reach extreme levels
of cleanliness.
1.4 Casting processes
In order to convert a liquid metal into a solid it is necessary to remove heat to (i)
cool the liquid below its melting point and (ii) remove the latent heat
(exothermic enthalpy release) during the transformation from liquid to solid.
This is carried out in a container called a mold.
In general, the transformation of an element of liquid into solid follows a heat
balance given by:
c
p
@T
@t
L
@f
s
@t
ÿ Q
S
V
1:93
1.7 Purification by zone melting (Pfann, 1952).
1.8 Schematic of concentration change during zone melting.
Descriptions of high-temperature metallurgical processes 27
where , c
p
and L are the density , specific heat capacity and volumetric latent heat
of the metal, respectively. S and V are the surface area and volume of the element,
and f
s
is the fraction solid. Q is the net heat flux to/from the volume element.
This equation can be closed by specifying an f
s
-t-T relat ionship if Q and f
s
can
be identified. Q is dependent upon the thermal field caused by the structure
around the element and the mold characteristics. The fraction solid, f
s
, depends
on the solidification kinetics, which is governed by the rates of nucleation and
growth. The specific conditions of the alloy chemistry and mold will govern
nucleation and growth rates and thermal fields, and this in turn will determine
solidification structure in terms of solid crystal size and shape, micro and macro
segregation and porosity.
Casting processes are carried out in a wide variety of mold types and
processes. They includes continuous processes such as:
1. Oscillating molds to cast profiles with uniform cross-sections such as billet,
blooms, slabs, thin slabs or rails. This is by far the dominant process for
producing steel bars and sheets. In this process, molten metal is continuously
poured through a submerged entry nozzle from a tundish into an oscillating,
water-cooled mold. A mold slag is used to cover the surface of the melt
meniscus and also to fill the gap between the mold and the solidifying shell.
The roles of the mold flux are to control heat transfer, lubricate the mold±
strand interface and to refine and protect the melt from impurities.
2. Moving molds using two large rotating wheels in between which thin
sections can be cast have resulted in the strip casting process that reduces to
a large extent the need for rolling in the production of sheet steels.
3. Atomization is a process by which metal powder is produced by inter-
cepting melt pouring from a tundish with a high velocity gas. By controlling
gas velocity and angle of intercept different shapes and sizes of powders can
be produced. Atomized powder is a major source for the powder metallurgy
industry.
Casting is also carried out in batch processes such as:
1. Large permane nt ingot molds. This is used to caste large parts that are to be
later machined into specific shapes and sizes. In the case of steels, this has
largely given way to continuous casting.
2. Permanent or expendable molds. This is used to shape-cast various
structural parts of complex geometry, e.g. in the automotive and aerospace
industries. Expendable molds include those made from sand or ceramics.
As mentioned before, the solidified microstructure depends on the cooling rate.
The cooling rate is determined by how fast heat is removed from the melt (heat
transfer). In casting processes, the heat transfer is thus the most pertinent
variable that needs to be controlled. During solidification, the melt adjacent to
the mold wall usually solidifies first since this is where it is coolest. The heat
28 Fundamentals of metallurgy
from the melt must thus be transported through (as shown in Fig. 1.9):
(a) the liquid metal;
(b) the solid metal shell that forms at the mold wall;
(c) across the interface between the solid shell and the mold; and
(d) the mold.
The rate of heat transfer depends on the design of the mold and thus the mold is
designed according to the desired microstructure and production rates in various
industrial casting processes. Heat is transferred by conduction, convection and
radiation. Usually the heat transfer through the liquid metal is fast due to the
high degree of convection in the melt.
The temperature profile and corresponding heat flux is calculated by simul-
taneously solving the multi-scale problem of heat transfer and solidification
kinetics. Solving a general coupled problem of solidification and heat transfer is
complicated and beyond the scope of this chapter. Two simplified cases, each
having some practical importance in the casting industry, are discussed below.
1.4.1 Heat conduction in the mold controls the heat transfer:
expendable ceramic molds (sand casting, investment
casting) used for shape casting
Molds made out of ceramics conduct heat so slowly that the entire temperature
drop occurs across the mold (i.e. temperature in solid and liquid metal and
mold±metal interface are equal and thus at thermal equilibrium). Furthermore,
the mold can often be assumed to be infinitely thick in such a batchwise casting
process. The situation is schematically shown in Fig. 1.10.
The Fourier's law solution to this problem yields the fol lowing expression for
the heat flux:
Heat flux q
k
mold
mold
C
p;mold
p
t
p
T
m
ÿ T
o
(1.94)
1.9 Schematic of the heat transfer resistances during solidification.
Descriptions of high-temperature metallurgical processes 29
The h eat generated per unit area during the growth of the solid is given by the
latent heat:
Released heat H
m
dx
dt
metal
M
W ;metal
(1.95)
The released heat is the heat that is to be transported in the heat flux and thus we
can equate the two: released heat heat flux. After integrating (with x 0 at
t 0) we end up with an expression for the solid thickness:
x
2T
m
ÿ T
o
M
W ;metal
metal
H
m
p
k
mold
mold
C
p;mold
p
t
p
1:96
1.4.2 Heat transport across the mold±solid metal controls the
heat transfer: strip casting, thin slab casting, wire casting,
die casting and atomization
This is often the case when casting thin cross-sections in short times using metal
molds. Mold thickness is usually large compared to the cross-section of cast
metal. This situation is shown in Fig. 1.11. In this case heat transfer in the ingot
and mold are so fast that it is the interfacial heat transfer that controls how fast
heat is transported. The heat transfer equations yield the following expressions
for the heat flux and shell thickness in this case:
1.10 Schematic temperature profile across a sand mold.
30 Fundamentals of metallurgy
q hT
m
ÿ T
o
1:97
x
hT
m
ÿ T
o
metal
H
m
M
W;metal
t 1:98
In the first equation, h is the interfacial heat transfer coefficient [J/(m
2
.K)]. Its
value depends on both the contact between the solid shell and the mold as well
as the surface properties of both and whether or not a gas fills the gap. Due to its
complexity, it must usually be determined exper imentally for each type of mold.
1.5 Thermomechanical processes
There are two basic goals in thermomechanical processing of metals:
1. Plastic deformation at ambient or elevat ed temperature of a cast metal into a
desired shape. This is necessary for metals that have not been cast into their
final shape such as continuously cast metal or ingots from electroslag
refining or vacuum arc re-melting. Alternative methods to achieve this are
shape casting (die or investment molds) or powder compaction.
2. Use of deformation and heat treatment to obtain micro-structural changes in
order to improve properties. This applies not only to mechanical properties
but also to others, e.g. electrical, magnetic and ferroelectric properties.
1.5.1 Deformation processing to obtain shape changes
The principle of deformation processing is to apply a mechanical force resulting in
stresses that exceed the metal's yield stress. For example, for the case of uniaxial
1.11 Schematic temperature profile across a metal mold and metal strip.
Descriptions of high-temperature metallurgical processes 31
tension, we can define a yield stress at the end of the Hooke's law regime. The
condition for reaching the yield stress is approximated through a suitable yield
criterion, e.g. Tresca or Von Mises. The metal flows due to the creation and
movement of dislocations. The yield stress increases as the deformation increases,
because more dislocations are created as they get entangled more into one another.
This causes an increase in the yield stress and is called work hardening:
o
(no work hardening) k* (
d
)
1/2
(1.99)
Here,
d
dislocation density.
Most deformation processe s can be described by the following characteristics
(shown in Fig. 1.12):
·
The metal has an initial shape and initial properties.
·
It goes through a plastic deformation zone.
·
The metal has a final shape and final properties.
·
The process is affected by the friction condition at the interface.
A few examples of deformation processes are listed below:
Rolling
Material is passed between two rolls of equal radius. This is used for reducing
the thickness of the plates and sheets (see Fig. 1.13).
Forging
A die is used to apply a force and change the shape of the metal sample. The
material is in compression during deformation (see Fig. 1.14).
Wire drawing
The material is drawn through a die and is in tension during the deformation (see
Fig. 1.15).
1.12 A generic schematic of a deformation process.
32 Fundamentals of metallurgy
1.13 Rolling.
1.14 Forging.
1.15 Wire drawing.
Descriptions of high-temperature metallurgical processes 33
Extrusion
Similar to wire drawing but material is in compression (see Fig. 1.16).
1.6 References
Alcock C.B. (1976) Principles of Pyrometallurgy, London, Academic Press, 180±226.
Arvanitidis I., Sichen D., Sohn H.Y. and Seetharaman S. (1997) `The Intrinsic Thermal
Decomposition Kinetics of SrCO3 by a Nonisothermal Technique', Metall Mater
Trans B, 28B, 1063±68.
Barksdale J. (1966) Titanium Its Occurrence, Chemistry and Technology, New York,
Ronald Press Co., 213±23, 400±79.
Bates P. and Muir A. (2000) `HIsmelt ± Low Cost Iron Making', Gorham Conference,
Commercializing New Hot Metal Processes ± Beyond The Blast Furnace, June 2000.
Biswas A.K. and Davenport W.G. (1976) Extractive Metallurgy of Copper, Oxford,
Pergamon, 81±82.
Bouris D. and Bergeles G. (1998) `Investigation of inclusion re-entrainment from the
steel±slag interface', Metall Mater Trans B, 29B, 641±649.
Cleaver J.W. and Yates B. (1973) `Mechanism of detachment of colloidal particles from a
flat substrate in a turbulent flow', J Coll Interface Sci, 44(3), 464±74.
Cramb A.W. and Jimbo I. (1988) `Interfacial considerations in continuous casting', W.O.
Philbrook memorial symposium, Toronto, Ontario Canada, 17±20 April 1988,
Warrendale, PA, USA, ISS, 259±71.
Danckwerts P.V. (1951) `Significance of liquid-film coefficient in gas adsorption', Ind
Eng Chem, 43, 1460±67.
Darken L.S. and Gurry R.W. (1952) Physical Chemistry of Metals, New York, McGraw-
Hill.
Evans J.W. and Koo C-H. (1979) `The Reduction of Metal Oxides', in Sohn H.Y. and
Wadsworth M.E., Rate Processes of Extractive Metallurgy, New York, Plenum,
285±321.
Gupta S. (1999) `Steelmaking-Technological Options', in International Conference on
Alternative Routes of Iron and Steelmaking (ICARISM `99), 15±17 September 1999,
Perth, WA, Australia, 81±84.
Habashi F. (1986) Principles of Extractive Metallurgy, Vol. 3 Pyrometallurgy , New York,
Gordon and Beach, 326.
1.16 Extrusion.
34 Fundamentals of metallurgy
Hojo J., Oku T. and Kato A. (1978) `Tungsten Carbide Powder Produced by the Vapor
Phase Reaction of the WCl
6
-CH
4
-H
2
System', J Less Common Metals, 59(1), 85±
95.
Hong L., Sohn H.Y. and Sano M. (2003) `Kinetics of Carbothermic Reduction of
Magnesia and Zinc Oxide by Thermogravimetric Analysis Technique',
Scandinavian J. Metallurgy, 32, 171±76.
International Iron and Steel Institute (2004) `World Steel in Figures', from website, http://
www.worldsteel.org/wsif.php.
Jones D. (1992) Principles and Prevention of Corrosion, 2nd edn, New Jersey, Prentice
Hall.
Kofstad P. (1988) High Temperature Corrosion, 2nd edn, New York, Elsevier.
Kopfle J., Anderson S.H. and Hunter R. (2001) `Scrap, DRI and Pig Iron: Whassup?' ISRI
Ferrous Conference, Chicago, IL, 2001; http://www.midrex.com/uploadedfiles/
Whassup.pdf.
Kozakevitch P. (1954) `Viscosity of blast-furnace slags', Revue Metall, 51, 569±87.
Kozakevitch P., Lucas L. and Louis D. (1968) `Part placed by surface phenomena in the
elimination of solid inclusions in a metal bath', Revue Metall, 65, 589±98.
Kozakevitch P. and Olette M. (1970) `Role of surface phenomena in the mechanism of
removal of solid inclusions', Production and application of clean steel, Intl.
Conference, 1970, Balatonfured, Hungary, London, The Iron and Steel Institute,
42±49.
Kozakevitch P. and Olette M. (1971) `Role of surface phenomena in the mechanism used
for eliminating solid inclusions', Revue Metall, 68, 636-46.
Lahiri A.K., Yogambha R., Dayal P. and Seetharaman S. (2002) `Foam in Iron and
Steelmaking', in Aune R.E. and Sridhar S., Proceedings of the Mills Symposium,
Metals, Slags, Glasses: High Temperature Properties and Phenomena, 22±23
August 2002, London, vol. 2, London, The Institute of Materials.
Lamprey H. and Ripley R.L. (1962) `Ultrafine Tungsten and Molybdenum Powders', J.
Electrochem Soc, 109(8), 713±15.
Machlin E.S. (1960) `Kinetics of vacuum induction refining: theory', Trans AIME, 218,
314±26.
MIDREX Technologies Inc. (2001) `2000 World Direct Reduction Statistics', http://
www.midrex.com/info/world.asp.
MIDREX Technologies Inc. (2004) `2003 World Direct Reduction Statistics', http://
www.midrex.com/info/world.asp.
Mills K.C. and Sridhar S. (1999) `Viscosities of iron and steelmaking slags', Ironmaking
Steelmaking, 26, 262±68.
Nakajima K. and Okamura K. (1992) `Inclusion transfer behavior across molten steel±
slag interfaces', 4th Intl. Conf. on Molten Slags and Fluxes, ISIJ, Sendai, 1992,
505±10.
Nikklason G.A. (1987) `Optical Properties of Gas-Evaporated Metal Particles: Effects of
a Fractal Structure', J Appl Phys, 62, 258±65.
Okamoto Y., Koyano T. and Takasaki A. (1987) `A Magnetic Study of Sintering of
Ultrafine Particles', Japan J Appl Phys, 26, 1943±45.
Otsuka K-I., Yamamoto H. and Yoshizawa A. (1984) `Preparation of Ultrafine Particles
of Nickel, Cobalt and Iron by Hydrogen Reduction of Chloride Vapors', J Chem Soc
Japan, 6, 869±78.
Padilla R. and Sohn H.Y. (1979) `Reduction of Stannic Oxide with Carbon', Metall Trans
Descriptions of high-temperature metallurgical processes 35