Mechanical Alloying Fundamentals and A (P R Soni
Mechanical Alloying Fundamentals and A (P R Soni
Mechanical Alloying Fundamentals and A (P R Soni
i
ii
MECHANICAL ALLOYING
Fundamentals and Applications
P.R. Soni
iii
Published by
Cambridge International Science Publishing
7 Meadow Walk, Great Abington, Cambridge CB1 6AZ, UK
http://www.cisp-publishing.com
P.R. Soni
Cambridge International Science Publishing
Conditions of sale
All rights reserved. No part of this publication may be reproduced or trans-
mitted in any form or by any means, electronic or mechanical, including pho-
tocopy, recording, or any information storage and retrieval system, without
permission in writing from the publisher
ISBN 1 898326568
iv
TO MY FATHER
Dr. P.R. Soni received his M.Sc. (1975) degree in Physics from the
University of Udaipur; M.E. (1979) and Ph.D. (1991) from the Uni-
versity of Rajasthan, Jaipur. Dr. Soni joined the Department of Metal-
lurgical Engineering at Malaviya Regional Engineering College, Jaipur
in 1980, where presently he is a Reader. He had been a Research
Fellow at the Indian Institute of Technology, Bombay, in 1989. Dr.
Soni was selected as Manager (R&D) of Nappro Synthetics in 1990.
For the last fifteen years, he has been actively engaged in re-
search and development activities related to mechanical alloying,
composite materials and other P/M materials, and is closely associ-
ated with industries through consultancy. He has published thirty tech-
nical papers in various peer reviewed journals. He is a member of
the APMI International and many other professional bodies.
v
vi
Contents
1 INTRODUCTION ......................................................................................... 1
1.1 HISTORY ........................................................................................................................... 1
1.2 BENEFITS OF MECHANICAL ALLOYING ............................................................... 3
vii
5.2.2 Nanocrystallization ....................................................................................................... 49
5.2.3 Solid Solubility Extension (SSE) ................................................................................. 51
5.3 ACTIVATION OF SOLID STATE CHEMICAL REACTION .................................. 51
viii
12.13 AMORPHOUS SOLIDS ........................................................................................... 136
12.14 NANOCRYSTALLINE MATERIALS .................................................................... 137
12.15 MECHANICALLY ACTIVATED CHEMICAL REACTIONS ........................... 138
12.16 OTHERS .................................................................................................................... 140
CONCLUSIONS .................................................................................................................. 140
LIST OF SYMBOLS ........................................................................................................... 145
INDEX .................................................................................................................................. 147
ix
PREFACE
From the total initial laboratory success in 1968, the process of me-
chanical alloying (MA) has been developed into a well-controlled pro-
duction operation over the years, and applied to develop varieties of
materials. Being a new field, there is a wealth of recent scientific lit-
erature available, but it is all scattered The problem of a beginner
to get started in a practical way with the MA technique. This book
tries to address this problem and is aimed at the undergraduates, post-
graduates, materials scientists and engineers who want to have in-depth
knowledge in this field. The book is also designed to serve as an in-
troductory and refreshment reference tool for the manufacturing en-
gineers actively involved in MA or the allied industry but are in need
of detailed knowledge of metallurgical engineering or materials science.
A two year metallurgical engineering or materials science course should
provide the necessary basis for comprehension of the material discussed
in the book.
This book tries to put forward the fundamentals of MA recipes where
the technique has been successful and highlights the areas in technology
where it can provide benefits in developing high-tech materials. Not
only this, many secrets of the MA processing approach are still in the
embryonic stage and this book creates a brainstorm in the mind of ma-
terials scientists and engineers to reveal the same. The book comprises
of twelve chapters. Starting from the historical development of the MA
technique and highlighting its benefits in Chapter 1, Chapter 2 discusses
the basic process of MA devices used for, and factors affecting the
process. This chapter deals with the different variations of MA which
have been developed over the course of time. Chapter 4 deals with
the process control agents used in MA, while Chapter 5 deals with
mechanisms involved in basic processes of MA, metastable phases for-
mation and activation of solid state chemical reactions. Chapter 6 deals
with energy transfer and energy maps in MA. Chapters 7 and 8 deal
with basics of consolidation of MA powders and mechanical proper-
ties of MA products, respectively. Chapter 9 explains the basics of
models developed to predict the results of an MA process. Joining tech-
niques for MA materials are discussed in Chapter 10. The two non-
equilibrium processing techniques, MA and rapid solidification, are com-
pared and combinations of the two to enhance the properties of the
xi
Mechanical Alloying
P.R. Soni
xii
Introduction
1 INTRODUCTION
1.1 HISTORY
The use of inert additions to improve elevated temperature mechani-
cal properties of metals was first exploited in 1910 by W.D. Coolidge
in thoriated tungsten [1]. The development of dispersion-strengthened
alloys by internal oxidation started in 1930 [2] and the invention of
dispersion-strengthened aluminium took place in 1949 [3]. However, the
relatively low melting point of aluminium was a severe limitation for
the use of SAP at elevated temperatures. This led to attempts in ap-
plying dispersion strengthening to higher melting point metals such as
copper and nickel.
In these metals, the self oxides cannot be used as they are not suf-
ficiently stable against Ostwald ripening at elevated temperatures. ThO 2-
dispersed nickel, having a finely distributed dispersoid, was produced
successfully by melting to improve the mechanical properties (1960)
[4]. The material has vastly developed elevated temperature properties.
However, the use of such materials was still limited due to their low
strength at intermediate temperatures and lack of corrosion resistance.
Hypothetically, these shortcomings could be solved by combining the
corrosion resistance and intermediate temperature strength of
precipitation-hardened, nickel-base superalloys with the high tempera-
ture strength and stability of oxide dispersion strengthening. One of the
primary hurdles to materialize this idea is the production of uniform
dispersion of fine oxide particles, less than 0.1 m in size, in alloy pow-
der particles in such a manner which leads to interparticle spacing of
less than 0.5 m in a consolidated product. The nickel-base superalloys
contain chromium, aluminium and titanium for effective precipitation
hardening, elements which are easily oxidizable. Aluminium and tita-
nium oxides are so stable that they cannot be reduced to a metallic
state, required in an alloy, without reducing deliberately the dispersed
oxide as well. Moreover, oxidation of these elements removes them as
a precipitation hardener in the alloy.
Four techniques, namely the simple mixing, ignition surface coat-
ing, internal oxidation and selective reduction, are available to com-
1
Mechanical Alloying
step impossible with gases, because of the stability of Al2O 3 and TiO 2.
These oxides can be reduced by the use of molten alkali and alkaline
earth metals. However, this introduces two major new problems, ex-
cessive growth of the dispersoid particles, and the necessity to remove
the reaction product oxides and carrier agent, usually salt.
To cope with these problems associated with the production of oxide
dispersion-strengthened nickel-base superalloys J.S. Benjamin (Father
of MA) and his colleagues at the Paul D. Merica Research Labora-
tory of the International Nickel Company processed a very high pu-
rity nickel powder, a fairly coarse chromium powder, a master alloy
powder of nickel, aluminium and titanium and a very fine powder of
Y 2O3 in a high energy attritor mill in the late 1960s [6]. As a historical
note, it should be recorded that at this stage, for the process of mixing/
milling the term mechanical alloying was coined by Ewan C.
MacQueen, a patent attorney for the International Nickel Company [7].
When such a mixture is mechanically alloyed, the degree to which its
various constituents maintain their form depends on their relative hard-
ness and their ability to withstand deformation. Nickel, which is the
softest constituent of the mixture, is the cemet that binds the other con-
stituents together. Chromium is somewhat harder and less ductile than
nickel, so it tends to form plate-like fragments that are embedded in
nickel. The Y 2O 3 disperses along the welds in the composite particles.
At the end of the process, the composite particle has a random dis-
tribution of oxide dispersoid in a metal-matrix composite. The powder
was then consolidated by hot extrusion. The grain size in the extruded
bar was very fine. However, to have maximum high temperature
strength, the grains should be coarser. For this, the extruded bar was
rolled and then annealed at about 1263C for 30 min, to recrystallize
to quite coarse grains.
As well as this, they could add elements such as tantalum, molyb-
denum and tungsten to the nickel-base superalloy, which gave added
strength at lower temperatures.
4
Introduction
References
1. W.D. Coolidge, Proc. Am. Inst. Elec. Eng., 961 (1910).
2. C.S. Smith, Min. and Met., 11, 213 (1930).
3. R. Irman, Tech. Rundschen (Bern), 36, 9 (1949).
4. G.B. Alexander, U.S. Patent 2, 972, 529 (1961).
5. J.S. Benjamin, Met. Trans., 1, 2943 (1970).
6. J.S. Benjamin, Scientific American, 235, 40 (4) (1976).
7. J.S. Benjamin, MPR, 45 (2), 122 (1990).
Questions
1. Which are the various techniques available for preparing dispersion
strengthened materials.
2. How is MA helpful in the case of reactive materials?
3. Who invented MA, when and where?
4. Give in brief, a chronological development of MA.
5. Enlist various advantages associated with MA.
5
Mechanical Alloying
2 MECHANICAL ALLOYING
2.1 ALLOYING MILLS
A variety of milling equipment such as attritor mills, vibratory mills,
high speed blenders and shakers, planetary mill, and even large diameter
conventional ball mills have been used to carry out MA. The milling
machine stresses the maximum number of individual particles in a
powder mass to undergo plastic deformation or initiate fracture with
a minimum of energy. The motion of the milling medium and the charge
varies with respect to the movement and trajectories of individual balls,
the movement of the mass of balls, and the degree of energy applied
to impact, shear attrition and compression forces acting on powder par-
ticles. Impact is the instantaneous striking of one object by another.
Both objects may be moving or one may be stationary. Attrition is the
production of wear debris or particles created by rubbing action be-
tween two bodies. This type of milling force is preferred when the
materials are friable and exhibit minimal abrasiveness shear consisting
of cutting or cleaving of particles, and is usually combined with other
types of force. Shear contributes to fracturing by breaking particles into
individual pieces, with the creation of a minimal of fines. Compres-
sion is the slow application of compressibility forces to a body. A choice
between these is likely to be determined by the end results required,
as well as the chemical and physical properties of the powder.
7
Mechanical Alloying
8
Mechanical Alloying
tion of the grinding balls works out to be 24g, where g is the gravi-
tational force [2].
Maximum milling action takes place adjacent to the chamber wall
and minimum at the centre, as the motion of the medium decreases
from the chamber walls to the centre of the mill tube. The entire charge
slowly revolves counterclockwise to the oscillatory vibrations and the
high g-forces experienced by the agitated balls leads to intense MA.
Vibratory mills utilize small size grinding balls because of higher im-
pact forces, frequencies and acceleration. In fact, it is the most en-
ergetic milling device as compared to other mills. These mills are ex-
cellent for MA of amounts of up to 4.5 kg or more, depending on the
apparent density of the powder. The vibratory mill has, however, not
found wide acceptance for bulk production.
High speed Spex shaker mills or blenders are ideally suited for labo-
ratory experiment. In fact, some of Benjamins experiments were also
carried out in a high speed shaker mill.
9
Mechanical Alloying
10
Mechanical Alloying
Fig.2.6 Regions of the attritor manifesting differing milling actions (Ref. 8, Met. Trans.,
24A, 175 (1993)).
These devices are not designed with specific needs of the MA proc-
ess in mind. This factor has become particularly obvious in recent
years, when the dependence of final products on the type of milling
device and energy regime used have been repeatedly reported. On these
premises a few improved milling devices have been proposed as dis-
cussed below.
11
Mechanical Alloying
12
Mechanical Alloying
Fig.2.9 Schematic of the ball milling with controlled ball movement: 1) rotating cell; 2)
balls; 3) magnets (Ref.10, Mat. Sci. and Eng., A134, 1356 (1991)).
Fig.2.10 Modes of operation of the ball mill: a) high energy mode, high rotation frequency;
b) low energy mode, low rotation frequency; c) high energy mode two point of equilibrium;
d,e) high energy mode, intermediate rotation frequency (Ref.10).
13
Mechanical Alloying
radius of the cell and Rb is the radius of the ball. Periodically, the outer
ball on the right-hand side gets released, completes most of the cir-
cle being pushed against the cell wall by the centrifugal force and hits
the left-most ball of the bottom. With this mode of operation, the
powder is worked by both by impact and shearing. The balls may be
confined to the bottom part of the cell for the whole time, either by
increasing the intensity of the magnetic field or by decreasing the fre-
quency c (Fig.2.10b). With this mode, the balls both rotate and os-
cillate around the equilibrium position of the bottom and the powder
is worked mostly by shearing.
In the third case, illustrated in Fig.2.10c, the ball movement caused
by the centrifugal force can be halted in two opposite directions: at the
lowest and highest point inside the cell. The ball trapped by the mag-
netic attraction in the upper position rotates with frequency b and can
be released to fall vertically on top of one of the bottom balls. The
two colliding balls rotate in opposite directions, which results in a com-
bination of shearing and uniaxial pressure at the surface of contact.
Two useful variations of mode (a) are shown in Figs.2.10c and 2.10d.
By slowing down the cell rotation frequency c a situation may be
achieved when the ball released from the bottom is not fully pinned
to the wall by the centrifugal force and can either hit one of the bottom
balls (Fig.2.10c) or the opposite cell wall (Fig.2.10d). The milling con-
ditions in situations (a), (d) and (c), (e) are similar, respectively.
In conventional ball mills, typical impact times are the order of
10 5 sec and the peak stress can reach 50 kbar. Depending on the col-
lision parameters, the orientation of the impact strain may be quite com-
plex and the temperature rise in the collision region may vary from a
few degrees to a few hundred degrees [11,12]. The mechanical energy
is utilized most efficiently if the balls collide head-on with maximum
possible velocities. Davice et al [12] reported eight head-on collisions
out of the total of 2132 collisions per second in a vibratory mill agi-
tated at 1200 rpm. There are only one to two collisions per second
in this device, but they are strictly head-on because of the guiding of
the ball movement by the magnetic field. The free-fall ball velocity dur-
ing an impact is 15 m/sec (it is increased in the presence of a mag-
netic field). For comparison, the average relative ball velocity in a vi-
brating mill varies from 3.9 to 6 m/sec [4], in a planetary mill [13]
it remains in the range of 2.54.7 m/sec and it is about 0.5 m/sec in
attritors [11]. The calculated values of collision time, Hertz radius (refer
to Section 9.2.3) and maximum impact stress have been found to be
6.510 5 sec, 4.610 4 and 37 kbar respectively for this mill. All these
values are close to the corresponding values quoted for commercial vi-
brating mills. Thus, the energy released per impact in this mill is not
14
Mechanical Alloying
16
Mechanical Alloying
17
Mechanical Alloying
Fig.2.13 Typical SEM layered structure of milled powders (Cr48% Nb elemental mixture)
indicating the degree of alloying as a function of milling time: a) 5 hrs; b) 15 hrs; c) 20
hrs; d) 25 hrs (A.H. Clauer and J.J. deBarbadillo (editors). Solid State Powder Processing
(1990), p.305 (The Minerals, Metals and Materials Society).
18
Mechanical Alloying
powder rises above a certain point, the cold worked metal particles may
undergo recovery and recrystallization. Therefore, water-jacketed milling
chambers are usually required for large, high energy vibratory and
attritor mills that reach temperatures as high as 200C [21].
19
Mechanical Alloying
bution of balls, even the shape of impellers in the case of attritor mills),
temperature, atmosphere and contamination.
2.3.1.4 Speed
The milling speed is one of the most important variables to be con-
sidered. Very low rotational speeds lead to very long periods of mill-
ing (>100 hrs) and a large inhomogeneity in the alloy because of in-
adequate kinetic energy input, resulting in insufficient localized attri-
bution heat input for alloying. Therefore, an extremely prolonged milling
time would presumably be required for homogeneous [20,36] alloying
(Fig.2.15). For speeds greater than the optimum, the milling time gets
reduced for the same number of revolutions and thus effectiveness of
alloying again decreases because of the decrease in the time available
for diffusion of the solute. However, very high speeds could lead to
excessive heating, high wear of the balls causing contamination from
the grinding medium and lower yields.
20
Mechanical Alloying
2.3.2 Temperature
The ambient temperature of MA and MG is an important parameter
which may influence the final structure. The amorphization of the sys-
tem, for which the heat of mixing is negative (H mix< 0) e.g. NiZr,
increase in ambient temperature increases amorphization due to the fast
diffusion rate of the constituent elements [27,37]. For the system, with
positive heat of mixing (H mix>0), the effect of ambient temperature
starts after a specific period of MA (e.g. CuTa, ambient temperature
effect starts after 30 hrs of MA). During this period, the alloying mostly
21
Mechanical Alloying
2.3.3 Atmosphere
Mostly, the MA and MG processes are performed in an inert atmos-
22
Mechanical Alloying
2.3.4 Contamination
The type and level of contamination from the wear of the mill chamber
and the grinding balls, though small (ppm level), may influence the
amorphization transformation path. For example, when MG of NiZr
is carried out in an attritor or vibratory mill, the compound transforms
directly to the amorphous state. But when the process is carried out
in a planetary mill, crystalline intermetallic compound appears as an
intermediate product which subsequently transforms to the amorphous
state. The difference is attributed to different levels of iron contami-
nation from the different mills [41]. There are cases when the change
of grinding medium balls (say from steel to WC) changes the end prod-
uct. The type of PCA used during MA influences the end product (refer
to Chapter 4), which may presumably be due to the level of carbon
added in the MA powder by the PCA.
References
1. W.E. Kuhn and M. Lucky, In: Fine Particles, W.E. Kuhn and J. Ehretsmann (eds).
The Electrochem. Soc. Inc. (1974), p.95.
2. G.B. Schaffer and P.G. McCormick, Mat. Forum, 16, 91 (1992).
3. W.E. Kuhn, In: Modern Developments in Powder Metallurgy, V.12, MPIF,
Princeton, N.J. (1980), p.195.
4. A.N. Patel and W.E. Kuhn, In: Modern Development in Powder Metallurgy,
V.13, MPIF, Princeton, N.J. (1980), p.2750.
5. B.N. Bachin, S.Ya. Kolupaeua, Yu.A. Kustor, A.J. Chernyok and B.V. Schetanov,
Poroshk. Metall., 235 (7) (1982), p.44
6. P.S. Gilman and W.E. Mattson, US Patent 4, 267, 959 (1986).
7. Anonymous, Attrition Mills, MPR, 51 (5), 58 (1996).
8. R.W. Rydin, D. Maurice and T.H. Courtney; Met. Trans., 24A, 175 (1993).
9. T.M. Cook and T.H. Courtney, Met. and Mat. Trans., 26A, 2389 (1995).
10. A. Calka and A.P. Radlinski, Mat. Sci. and Eng., A134, 1356 (1991).
11. D.R. Maurice and T.H. Courtney, Met. Trans., 21A, 289 (1990).
12. R.M. Davice, B. McDermott and C.C. Koch, Met. Trans., 19A, 2867 (1988).
13. J. Eckert, L. Schultz and E. Heustern, J. Appl. Phys., 64, 3224 (1988).
14. A.P. Radlinski and A. Calka, Mat. Sci. and Eng. A134, 1376 (1991).
15. A.P. Radlinski, A. Calka and B.W. Ninham and W.A. Kaczmarek. Mat. Sci. and
Eng. A134, 1346 (1991).
23
Mechanical Alloying
16. R.B. Schwarz and C.C. Koch, Appl. Phys. Lett., 49, 146 (1986).
17. J.S. Benjamin, Scientific American, 234, 40 (1976).
18. ASM Metals Handbook, Ninth Edition, V.7, 1990, p.56.
19. J.S. Benjamin, Mat. Sci. Forum, 88-90 (1992).
20. E.S. Rao, In: Powder Metallurgy - Recent Advances, V.S. Arunachalam
and O.V. Roman (eds), Oxford IBM, New Delhi (1989), p. 27.
21. H. Kimura, M. Kimura and F. Takada, J. Less-Common Metals, 140, 113 (1988).
22. J. S. Benjamin and T.S. Volin, Met. Trans, 5A, 1929 (1974).
23. K.H. Karmer, Powd. Metall. Int., 9, 105 (1977).
24. P.S. Gilman and W.D. Nix, Met. Trans., 12A, 513 (1983).
25. T.K. Wassel and L. Himmel, TACOM-TR-12571 (1981).
26. E. Bonetti, G. Cocco, S. Enzo and G. Valdre, Mat. Sci. and Tech., 6 (12), 1258
(1990).
27. L. Guoxian, W. Erde and W. Zhongren, J. of Mat. Proc. and Tech., 51, 122 (1995).
28. C. Michealsen and E. Hellstern, J. Appl. Phys., 62 (1), 117 (1987).
29. L.M. Di, PI. Loeff and H. Bakker, Mat. Sci. and Eng., A134, 1323 (1991).
30. B.Q. Li and Y.N. Wang, J. Appl. Phys., 75 (3), 1783 (1994).
31. H. Hashimoto and R. Watanabe, Met. Trans. JIP, 31 (3), 219 (1990).
32. E. Gaffet, Mat. Sci. and Eng., 132 (2), 181 (1991).
33. K.B. Gerasimov, J. Mat. Sci., 26 (9), 1296 (1991).
34. P. Yang, J. Jpn. Soc. Powd., Powd Met., 37 (5), 623 (1990).
35. A. Lasonna and M. Magini, Acta Met., 44 (3), 1109 (1996).
36. E.S.B. Rao, R.M. Mallya and D.H. Sastry, Trans. IIM, 39, 596 (1986).
37. C.H. Lee, T. Fukunage and U. Mizutani, Mat. Sci. and Eng.., A134 1334 (1991).
38. C.H. Lee, T. Fukunage and U. Mizutani, Jpn. J. Appl. Phys., 29, 540 (1990).
39. C.C. Koch, D.B. Cavin, C.G. McKamey and J.O. Scarborough, Appl. Phys. Lett.,
43 1017 (1983).
40. K.Y. Wang, T.D. Shen, M.X. Quan and J.T. Wang, J. Mat. Sci. Lett., 11, 129
(1992).
41. A.W. Weeber, W.J. Haag, A.J. H. Wester and H. Bakker, J. Less-Common Metals,
140, 119 (1988).
Questions
1 Which are the mills in practice for MA?
Which mill is most energetic amongst them?
2. Why, in general, are MA mills energy inefficient.
3. Describe an attritor mill, explaining its salient features.
4. Suggest measures to improve efficiency of an attritor mill.
5. What is a uni-ball mill? Explain specific advantages associated with
it?
6. Why is a vibratory mill preferred for amorphization.
7. Give schematic of a planetary mill.
8. Write a brief note on grinding medium used for MA.
9. What advantages are associated with large diameter horizontal ball
mills used for mechanical alloying?
10. How is the use of grinding balls of different diameter helpful in
improving efficiency of an attritor mill?
11. Explain how the use of a uni-ball mill can avoid use of PCAs in the
case of ductile-ductile system?
12. Define MA. How is it different from MG?
24
Variations of Mechanical Alloying
3 VARIATIONS OF MECHANICAL
ALLOYING
3.1 REACTION MILLING
While most MA is conducted in an inert atmosphere, there can be ad-
vantages to MA in a reactive atmosphere. In reaction milling, metal
powders react extensively with the milling fluid that is reactive dur-
ing milling. For example, processing powders in nitrogen atmospheres
can result in the formation of nitrides or nitrogen can be trapped in
the metal matrix [1,2,3,4], while processing in hydrogen atmospheres
may form metastable hydrides [5].
The metal powder is comminuted to a particle size much smaller than
the starting one. In general, chemical reactions between the fluid and
the powder assist comminution of metal powders by not allowing them
to weld together or to the balls/chamber walls.
Reaction milling can also be used to obtain the desired dispersoid
by introducing an element that reacts during milling or during subse-
quent heat treatment, or in part during both. For example, MA of Al
and C (lamp black or graphite) results in an AlAl4C 3 composite powder
[6], and of AlBgraphite in an AlBC composite powder [7].
The process can also be used to produce nanostructured and amor-
phous materials, if milled for prolonged hours [1], e.g. when iron pow-
der is mechanically alloyed with nitrogen at room temperature, it re-
sults in a highly disturbed, non-equilibrium, microfine microstructure
with an enhanced nitrogen concentration entrapped in both the metal
matrix and the highly disturbed regions. During MA, the impact deforms
the particles plastically, creating a new surface. The newly exposed sur-
face is highly reactive to the nitrogen atmosphere and gets adhered to
the fresh surface, dissociates and subsequently is incorporated into the
matrix when the particles become cold welded. The nitrogen infusion
by MA results in a linear increase in nitrogen concentration for up to
100 hrs and a nitrogen level of up to 1.0% in pure iron (the interstitial
nitrogen solubility in bcc iron is less than 0.05%) (Fig.3.1). During
subsequent consolidation, the matrix nitrogen diffuses rapidly and be-
come trapped in the heavily deformed microstructure and at the subgrain
25
Mechanical Alloying
boundaries. This entrapped nitrogen retards the grain growth and leads
to a nanostructural material.
3.2 CRYOMILLING
With the recognition of the fact that cryogenic temperatures (which are
used to control homologous temperatures) can be used to control the
MA process [8], the concept of cryomilling came into existence. In
cryomilling, liquid nitrogen is added directly to the milling vessel and
the technique has been applied to systems ranging from aluminium to
nickel-base superalloys [9,10]. In the case of aluminium, extremely fine
Al(ON) particles are reported to be formed in addition to the aluminium
oxide present on the surface of the starting powder particles. Cryomilling
of Al50 at% Ti in liquid nitrogen results in the formation and decom-
position of B2 TiAl, and the formation of AlN and Al 2O 3 , which
impede the grain growth [11].
26
Variations of Mechanical Alloying
d = d 0 (1/a) n (3.1)
where d0 is the initial particle diameter and 1/a is the reduction by rolling
of one cycle. However, due to changes of the stainless tube after each
rolling cycle, the kneading of powders may deviate to some extent from
the relationship given in Eq. (3.1). The sharp decrease in reduction of
particle size after several initial rolling cycles (Fig.3.3) suggests a
change in the kneading mechanism, probably by clipping of particles.
P.H. Shingu et al [14,15] have employed this technique for four binary
systems; AgFe, CuFe, AgCu and AlFe. The heat of mixing for
these alloy systems varies from large positive (AgFe) to moderately
large negative (AlFe). It was shown that due to the simple kneading
of elemental powder mixtures by the repeated rolling, metastable al-
loy phases such as nanocrystalline structure, supersaturated solid so-
lution and amorphous phases can be formed. These results indicate the
impact force present in an attritor or ball mill may be effective but not
the essential factor for solid state alloying. Thus, this technique can
further aid in understanding of the MA process.
27
Mechanical Alloying
Fig.3.4 Flow sheet of the MA process, with two possible routes (Ref.17).
28
Variations of Mechanical Alloying
Thermally
MA1 MA2
treated
Oxygen ( %) 1
References
1. J.C. Rawers and R.C. Doan, Met. and Mater. Trans., 25A 381 (1994).
2. M. Miki, T. Yamasaki, Y. Ogino, Mater. Trans. JIM, 34 (10) (1993) p.952.
3. H. Yasuda, Mater. Sci and Eng., A159, 676 (1994).
4. T.D. Chem, and C.C. Koch, Nanostruct. Mater., 5 (6), 615 (1995).
5. M. Baricco, J. Non-Crystalline Solids, 155, 156 & 527 (1993).
6. H. Danninger, G. Jangg and J. Zbiral, In: Solid State Powder Processing, H. Clauer
and J.J. deBarbadillo (eds), The Minerals, Metals and Materials Society,
Warrendale, PA (1990), p.241.
7. T. Takahashi and M. Motoyama, Preparation of particle dispersion strengthened
aluminium by MA, Presented at Powder Metallurgy World Congress, San Francisco,
CA, 21-26 Jan, 1992.
8. J.S. Benjamin and M.J. Bomford, U.S. Patent 3, 816, 080 (1974).
9. R.P. Luton and J. Valone, U.S. Patent 4, 647, 304 (1987).
10. M.J. Luton, Symp. Proc., Mater. Res. Soc. L.E. McCandlish (ed), Pittsburgh, PA,
132 79 (1989).
11. M.J. Luton, Nanostruct. Mater., 5 (6), 631 (1995).
12. M. Atzmon, K.M. Unruh and W.L. Johnson, J. Appl. Phys., 58, 3865 (1985).
13. F. Bordeau, A.R. Yavari and P. Desre, Mat. Sci. and Eng., A197, 129 (1988).
14. P.H. Shingu, K.N. Ishihara, K. Venishi, J. Kuyama, B. Huang and S. Nasu., Ibid
ref.6, p.21.
29
Mechanical Alloying
15. K. Uenishi, K.F. Kobayashi, K.M. Ishihara and P.H. Singhu, Mat. Sci. and Eng.,
A134, 1342 (1991).
16. P. Le Brun, L. Froyen and L. Delaey, Mat. Sci. and Eng., A157, 79 (1992).
17. X. Niu, P. Le Brun, L. Froyen, C. Peytour and L. Delaey, Powd. Met. Int., 25,
(3), 119 (1993).
18. G.B. Schaffer and P.G. McCormick, Met. Trans., 22A 3019 (1991).
19. M.A. Zaidi, J.S. Robinson and T. Sheppard, Mat. Sci. and Tech., 1 737 (1985).
20. P. Le Brun, X. Niu, L. Froyen, B. Munar and L. Delaey, Ibid ref.6, p.273.
21. J. Kihara, Productive Mechanical Alloying by Repeated Powder Forging, Presented
at the 1996 World Congress on Powder Metallurgy, Washington DC, WA, June,
1996.
Questions
1. Why are PCAs required during MA?
What other benefits are associated with the use of PCAs?
2. What are the commonly used PCAs in MA?
3. How PCAs affect MA powder hardness?
4. Why is the role of PCA most critical during the earliest stage of
milling?
5. What factors decide the amount of PCAs being used?
6. Why is degassing necessary for ductile-ductile MA powders?
7. What is the role of oxygen to carbon ratio in organic PCAs.
8. What contamination can PCAs cause? How can this contamination
be avoided?
9. In what way can excess use of PCAs be harmful?
10. How use of PCAs affects morphology of powder produced.
11. Give two examples to cite the fact that use of PCAs may lead to
structural changes in the powder.
30
Process Control Agents in Mechanical Alloying
Composition ( %)
PCA Chemical Formula
C O H N
Ethylene bis disteramide C2H 22(C18H 36ON) 77.3 5.4 12.4 4.8
(Nopcowax-22 DSP)
31
Mechanical Alloying
Table 4.2 Effect of PCA identity on powder size and shape (Ref.4)
PCA
PCA oxygen/carbon Powder size Powder shape
ratio
32
Process Control Agents in Mechanical Alloying
Fig.4.2 Effect of PCA identity and milling time on welding (elemental Al2% Cu alloy)
(Ref.4).
Fig.4.3 (right) Effect of PCA identity and level on PCA powder hardness (elemental
Al2% Cu alloy) (Ref.4).
33
Mechanical Alloying
in the composition being milled may react with such contaminates [7].
Organic PCAs also form hydrogen during MA which remains in the
granulates, necessitating a degassing step. In an effort to reproduce
high purity MA material, Gualandi [8] altered this practice by using sili-
con grease.
References
1. P.S. Gilman and W.D. Nix, Met. Trans., 12A, 813 (1981).
2. A.P. Radlinski, A. Calka, B.W. Ninham and W.A. Kaczmarek, Mat. Sci. and Eng.,
A134, 1346 (1991).
3. J.S. Benjamin and M.J. Bomford, U.S. Patent 3.816, 080 (1974).
4. J.H. Weber, In: Solid State Powder Processing, A.H. Clauer and J.J. deBarbadillo
(eds), The Minerals, Metals and Materials Society, Warrendale, PA (1990), p.227.
5. W.A. Kaczmarek, R. Bramley, A. Calka and B.W. Ninham, In: Conf. Proc. Intermag
90, Brighton, UK, April 1990.
6. F. Faudot, J. Mat. Sci., 28, 2669 (1993).
7. M. Nose, J. Jpn. Soc. Powd. Met., 42 (2), 166 (1995).
8. D. Gualandi and P. Johenson, In: Modern Developments in Powder Metallurgy,
H.H. Hausner (ed.) V.3, Plenum Press, New York (1966), p.36.
Questions
1. Describe MA by repeated rolling.
2. What is cryomilling?
3. For which various purposes can reaction milling be used?
4. How is reaction milling able to extend nitrogen solubility in iron?
5. What is DMA? How is it beneficial?
34
Mechanisms in Mechanical Alloying
5 MECHANISMS IN MECHANICAL
ALLOYING
The MA technique is being used mainly for three types of process-
ing: alloying, metastable phase formation and activation of chemical re-
actions. The underlying mechanisms involved in these processes to the
extent known today are discussed here.
5.1 ALLOYING
To discuss the mechanism of MA, it is convenient to divide the powder
charge into three systems:
ductileductile
ductilebrittle
brittlebrittle
35
Mechanical Alloying
37
Mechanical Alloying
Fig.5.3 Lattice parameter vs milling time for silicon and germanium powder for the
composition Ge72 at.% Si (Ref.6, Met. Trans., 19A, 2867 (1988). (The Minerals, Metals
& Materials Society)).
Fig.5.4 (right) SEM micrograph of interparticle 'necking' in Ge72 at.% Si after milling
for 8 hrs (Ref.6, Met. Trans., 19A, 2867 (1988)).
38
Mechanisms in Mechanical Alloying
40
Mechanisms in Mechanical Alloying
Amorphization by MA
Several theories have been suggested for the mechanism of the amor-
phization phase formation by MA. These include:
41
Mechanical Alloying
Table 5.1 Typical list of amorphous alloys formed by MA (Ref.22 and 29).
Hmix Atomic
Alloy A 1xBx
(kJ/mole)
(kJ/mol) size ratio
HfAl 40 1.17
HfCu 23 1.30
HfNi 44 1.35
NbCuGe 2 1.14
NbCuSi 2 1.14
NbNi 32 1.18
PdSi 37
SnNi 4 1.31
SnNb 5 1.11
TiCu 18 1.15
TiNi 39 1.19
ZrCo 42 1.28
ZrFe 26 1.27
ZrNi 51 1.29
ZrV 4 1.19
CuTa +3 0.870
CuV +7 0.950
NiW 5 0.855
CuAl 2 0.892
TaAl 46 1.02
NbAl 44 1.03
TiNiCu
AlNiFeGd
42
Mechanisms in Mechanical Alloying
44
Mechanisms in Mechanical Alloying
heat of mixing [27] and therefore the stored energy of cold work is
believed to play only a minor role. It is well-known, however, that SSAR
does not take place in bi-layers made of defect free single crystals and
that the nucleation of amorphous phase usually starts at an interface
near a triple point defect [20]. The presence of structural defects which
are thermally stable at the temperature where SSAR takes place, is there-
fore a necessary condition in order for the amorphization to proceed
in mechanical alloying.
However, this SSAR model based on amorphization in thin films does
not explain the amorphization in systems with H mix>0. In the case of
AB thin film multilayers, it has been found experimentally that com-
pound nucleation and growth take place when the amorphous layer
thickness grows beyond a critical thickness x c. The growth in the critical
thickness x c has been correlated with the ratio of diffusivities of the
D
two species ( xc a B ). A.R. Yavari et al [28] reviewed this for the
DA
MA case in the light of the fact that the thermodynamic driving force
for crystallization disappears with the large concentration gradients is
shown to depend on the heat of mixing H mix . Neither H mix < 0 nor
D B >>D A are required for amorphization by MA. The critical thickness
x c is shown to be 2nm for the alloys with H mix=0, such that amor-
phization by solid-state remains insignificant (in the case of thin films).
It is shown, however, that in MA, amorphous layer growth is accom-
panied by a simultaneous thickness shrinkage due to mechanical de-
formation. This allows the amorphous interlayer thickness to remain
below the critical value x c whilst its volume fraction increases.
In another approach to explain the SSAR by MA, Y. Chakk et al
[29] developed an atomistic model based on the assumption that
amorphization is obtained when a solute atom penetrates into the in-
terstitial sites and distorts the lattice. When such distortions, even when
the chemical diffusion is of less significance (i.e. H mix>0 for systems
like CuTa), reach some critical value, the long-range order of the lattice
is destroyed and the amorphous phase is obtained. It was established
that for such a mechanism to be operative, lower and upper limits of
the atomic size ratio (ASR) and the minimum solute concentration can
be established (refer to Section 9.3).
This idea is demonstrated in Fig.5.9. It is divided into four regions.
The first region (I) includes systems which are characterized by a low
negative enthalpy of mixing and a small difference in atomic radii, as
well as pure metallic elements with H mix = 0 and r A/r B = 1. In this
region, the systems are not amenable to amorphization by MA. The
second region (II) corresponds to a low negative enthalpy of mixing
45
Mechanical Alloying
but a large deviation from unity of the ASR. In this region all the sys-
tems were successfully amorphized by MA and those systems having
low negative or positive heat of mixing could not be amorphized in dif-
fusion couples. The third region (III) contains systems which are char-
acterized by a high negative enthalpy of mixing and the ASR is close
to unity. In this region, the systems were amorphized by MA. The forth
region (IV) is characterised by a high negative enthalpy of mixing and
a large atomic size mismatch. In this region the systems were
amorphized by both MA and SSAR of diffusion couples.
The ASR criterion should be regarded as a preliminary step for
predicting the ability to amorphize a bi-elemental system by MA. The
next step was to determine the atomic concentration range where the
amorphization can be observed.
The maximum strain energy is found to be only about 2% of the
total energy stored and reveals that most of the energy of the cold work
must be stored in the grain boundaries (of less mobile species) and not
in the lattice strain [30]. However, the strain energy has been found
to represent about half of the enthalpy of crystallization of the amor-
phous alloy powder. It therefore can contribute significantly to the
amorphization reaction (however, it represents only 8% of that of fusion
and 4% of the heat of mixing). Also the rate of amorphization has been
found to be maximum when strain is maximum, it is believed that lattice
strain plays an important role in the amorphization reaction.
Amorphization by MG
Amorphous alloy powders have also been prepared by MG of single
phase intermetallic powder [30,31], mixtures of intermetallic powders
[19] (Ni 32TiNi 45Nb55, NiTi 2NiNb) and mixtures of intermetallic and
elemental powders [32]. (It is surprising that the same amorphous alloy
46
Mechanisms in Mechanical Alloying
47
Mechanical Alloying
LM bI / I g
s t t
p
OP 0. 5
S=
MN bI / I g
s t t =0
p
PQ (5.1)
Transformation path
Kim and Koch [37] reported that during the MA of niobium and tin
powders in a molar ratio 3:1, the Nb 3Sn intermetallic (A15 structure)
forms first and with continued ball milling, transforms to the amorphous
Nb 75 Sn 25 alloy. A direct amorphization reaction was also observed by
Tiainen and Schwarz during the ball milling of pure nickel and tin pow-
ders [38]. Likewise, several amorphous alloy powders have been pro-
duced by MA. It is thus logical to ask whether during the MA of a
mixture of pure powders A and B, the amorphous alloy (AB) am forms
by the direct reaction
( A+B) (AB) am
48
Mechanisms in Mechanical Alloying
5.2.2 Nanocrystallization
When nanocrystallization in pure metals is carried out by MA, the heat
enthalpy (fusion) (625%) and heat capacity (814%) are raised due
to the high density of grain boundaries and the incorporation of lat-
tice defects into the crystal, which results in lattice softening and in-
creased unharmonicity in the lattice [40]. The enthalpies stored due to
extensive ball-milling are considerably higher than for other known cold
working methods of metals and alloys, which are rarely found to ex-
ceed 1 to 2 kJ/mol and are never more than a small fraction of the
heat of fusion H f [41]. In the case of the ball milling process, the
enthalpy determined is considerably larger and can reach values typical
for crystallization enthalpies of a metallic glass (0.4 Hf ). The maxi-
mum dislocation densities measured in heavily deformed metals are less
than 10 13 cm 2, which would correspond to an energy of less than 1
kJ/mol. Typical energies determined experimentally by differential thermal
calorimetry for equilibrated high-angle grain boundaries of high melt-
ing point metals are 10 4 Jcm 2 [42]. Assuming that the two top
monolayers of the subgrains belong to a grain boundary, about 50% of
the atoms of a nanocrystalline material with an average grain size
boundary thickness of 15 to 20 corresponds to an energy of roughly
70 J/cm 3 or 1 kJ/mol. As a consequence, the grain boundary energy
of the ball-milled nanocrystalline solids seems to be considerably larger
than the grain boundary energy of fully equilibrated grain boundaries.
The process of the grain size reduction can be understood in the
following way. In general, plastic deformation proceeds by slip and
49
Mechanical Alloying
Fig.5.12 Atomic resolution TEM bright-field image including the corresponding diffraction
pattern of Fe powder after 24 hrs milling (Ref.40, Met. Trans., 21A, 2333 (1990)).
twinning at low and moderate strain rates, whereas at high strain rates,
the formation of shear bands, which consist of a dense network of dis-
locations, becomes the dominant deformation mechanism. These shear
bands in which the deformation is localized have a typical width of 0.1
to 1 m [2]. The local shear instability of the crystal lattice may be
triggered by material inhomogeneities and is probably enhanced by ther-
moplastic instabilities due to non-uniform heating in certain regions during
the deformation process. In the early stage of ball-milling, the aver-
age atomic level strain increases due to the increasing dislocation density.
At a certain dislocation density, within these heavily strained regions,
the crystal disintegrates into subgrains which are initially separated by
low-angle grain boundaries with offset angles of less than 20. Dur-
ing processing, deformation occurs in shear bands located in previously
unstrained parts of the material. The size of the subgrains in the ex-
isting band is further reduced to the final grain size, and the relative
orientation of the subgrains with respect to each other ultimately be-
comes completely random (Fig.5.12).
The nanocrystalline grains themselves are relatively dislocation free,
as suggested by several factors [43]. Once an entirely nanocrystalline
structure is achieved, further refinement seems to be impossible. The
very high stresses required for dislocation movement hinder plastic
deformation of very small crystallites. Hence, further deformation and
energy storage can only be accomplished by a glide along the grain
boundary, which results in a random rotation of the subgrains.
In the case of an alloy system, the presence of the alloying element
[44] plays an important role to obtain nanosize crystals. As shown in
Fig.5.10 the strain in the material increases considerably during the mill-
50
Mechanisms in Mechanical Alloying
ing. The major contribution to the lattice strain comes from the high
density of dislocation [27]. The final fracture of the crystal will probably
occur by a shear in this highly distorted region of the crystal.
The fact that even for extended milling the elemental powders ex-
hibit a larger grain size and a smaller r.m.s. strain than the interme-
tallic compound [45] suggests that nanocrystalline materials are more
easily obtained for alloys than for the pure metals.
Thus, it can be said that, if from the thermodynamics point of view
the amorphous phase is significantly less stable, the desired
nanocrystalline phase can be formed by MA. In particular, for inter-
metallic compounds, the rise in free energy due to chemical disorder
during milling has to be taken into account. Therefore, intermetallic
compounds exhibiting a large difference in the free energy with respect
to the amorphous phase are most favourable for the preparation of
nanocrystalline microstructures. A transformation into a nanocrystalline
metastable solid solution can occur if this phase has a higher stabil-
ity than the amorphous phase, as has been demonstrated in the case
of NbAl [45].
However, it should be realized clearly that nanocrystalline structures
do not form in alloy systems within the field of complete solid solu-
bility. In such cases, a fine lamellae structure is produced by MA. By
varying the composition and milling conditions, either an equiaxed crys-
talline structure with grain size less than 10 nm or an amorphous struc-
ture results. Nanocrystalline cermets can be produced by MA of metallic
powders and metalloids. In mutually immiscible, e.g. AgFe, systems
MA results in nanocrystallization with the extension of solid solubil-
ity rather than amorphous phase.
Fig.5.13 Schematic showing variation in the critical reaction temperature maxima with
milling time (Ref.50, Met. Trans., 23A, 1285 (1992) (The Minerals, Metals and Materials
Society)).
52
Mechanisms in Mechanical Alloying
bation milling time tig required for the ignition temperature of the powder
mixture Tig to be reduced to a level equal to that achieved locally during
collision events. By changing the milling parameters (i.e. milling speed,
BPR, ball size), tig the incubation period can be changed but the critical
temperature and the ignition temperature remain unaffected for a system
[50].
References
1. J.S. Benjamin and T.E. Volin, Met. Trans., 5, 1930 (1974).
2. E. Hellstern, H.J. Fecht, C. Garland and W.L. Johnson, MRS Symp. Proc., 132,
(1989) 137.
3. J.S. Benjamin and M.J. Bomford, Met. Trans., 7A, 1301 (1977).
4. J.H. Weber, SAMPE Qly., 11 35 (1980).
5. G.B. Schaffer and P.G. McCormick, Met. Trans., 21A, 2787 (1990).
6. R.M. Davis, B. McDermott and C.C. Koch, Met. Trans., 19A, 2867 (1988).
7. P.J. Miller, C.S. Coffey and V.F. Devart, J. Appl. Phys., 54, 913 (1986).
8. H.M. McMillan and N. Gane, J. Appl. Phys., 41, 672 (1970).
9. P.W. Bridgman, Phys. Rev., 48, 825 (1935).
10. P.W. Bridgman, J. Appl. Phys., 18, 246 (1947).
11. J.S. Benjamin, Scientific American, 234, 40 (1976).
12. S.K. Kang and R.C. Benn, Met. Trans., 18A, 747 (1987).
13. T.K. Wassel and I. Himmel, Tr - 12571 (US Army Automotive Command, R and
D Control, Michigan) May 1981.
14. B.T. McDermott and C.C. Koch, Scripta Metall., 20, 669 (1986).
15. R.L. White and W. Nix, In: Proc Conf. New Developments and Applications in
Composites. K. Wilsdorf and W.C. Harrington Jr. (eds)., The Met. Sco. AIME,
N.Y. (1978).
16. E.S.B. Rao, R.M. Mallya. D.H. Sastry and V.S. Arunachalam, In: Proc. Int. Conf,
Horizons in Powder Metallurgy, W.A. Kaysser and W.J. Huppman (eds) Verlog
Schmid GAAH, Freiburg/Germany (1988), p.1275.
17. E.S.B. Rao, In: Powder Metallurgy - Recent Advances, V.S. Arunachalan and
O.V. Roman (eds), Oxford and IBM, New Delhi (1989), p.27.
18. C.C. Koch, O.B. Cavin, C.G. McKamey and Y.O. Scarborough, Appl. Phys. Lett.,
43, 1017 (1983).
19. R.B. Schwarz and C.C. Koch, Appl. Phys. Lett., 49, 146 (1986).
20. R.B. Schwarz, R.R. Petrich and C.K. Saw, J. Non-Crystalline Solids, 76, 281
(1985).
21. W. Krauss, C. Politis and P. Weimar, MPR., 43 (4), 231 (1988).
22. R.B. Schwarz and P. Nash, J. of Metals, 91, 27 (1989).
23. R.B. Schwarz and W.L. Johnson , Phys. Rev. Lett., 51, 415 (1983).
24. K.F. Kobayashi, N. Tachibana, P.H. Shingre, J. Mat. Sci., 25, 3149 (1990).
25. N. Saunders and A.P. Miodownik, J. Mater, Res., 1, 38 (1986).
26. Z.H. Yan, M. Oehring and R. Bormann, J. Appl. Phys., 76 (2), 2498 (1992).
27. E. Hellstern, H.J. Fecht, Z. Fu and W.L. Johnson, J. Appl. Phys., 65, 305 (1989).
28. A.R. Yavari and P.J. Derse, Mat. Sci. and Eng., A 134, 1315 (1991).
29 Y. Chakk, S. Berger, B.Z. Weiss and E.B. Levinson, Acta Metall.and Mater., 42
(11), 3679 (1994).
30 A.Ye Yermakov, Ye.Ye. Yurchikov and V.A. Barinov, Phys. Met. Metall., 52, 50
(1981).
31. Ibid ref. 30, 53, 935 (1982).
32. P.Y. Lee, J. Jang and C.C. Koch, In: Solid State Amorphization Transformations,
53
Mechanical Alloying
R.B. Schwarz and W.L. Johnson (eds), Elsevier, Lausanne (1988), p.28.
33. C.C. Koch and M.S. Kim, J. Phys. (Paris) Colloq., 46, 8-573 (1985).
34. R.B. Schwarz and R.R. Petrich, J. Less-Common Metals, 140, 99 (1988).
35. C.C. Koch, In: Solid State Powder Processing, A.H. Clauer and J.J. deBarbadillo
(eds), The Minerals, Metals and Materials Society, Warrendale, PA (1990), p.35.
36. G.J.C. Carpenter and E.M. Schulson, J. Nuclear Mat., 23, 180 (1978).
37. M.S. Kim and C.C. Koch, J. Appl. Phys., 62, 3450 (1987).
38. T.J. Tiainen and R.B. Schwarz, Ibid Ref. 32, p.69.
39. R. Bormann and R, Busch, In: Proc Conf, New Materials by Mechanical Alloying
Technique, E. Arzt and L. Schultz (eds), Deutsche Gesellschaft fr Metallkunde,
Germany (1989), p.73.
40. H.J. Fecht, E. Hellstern, Z.F, and W.L. Johnson, Met. Trans., 21A, 2333 (1990).
41. W.L. Johnson, Prog. Mater. Sci., 30, 81 (1986).
42. W. Gust, B. Predel and K.J. Stenzel, Z. Metallkd, 69, 721 (1978).
43 D.A. Rigney; Annu, Rev. Mater. Sci., 18, 141 (1988).
44. M.L. Trudeau and R, Schultz, Mat. Sci. and Eng., A134, 1361 (1991).
45. M. Oehring and R. Bormann Mat. Sci. and Eng., A134, 1330 (1991).
46. G.B. Schaffer and P.G. McCormick, Met. and Mater. Trans., 21A, 2789 (1990).
47 G.B. Schaffer and P.G. McCormick, Met. and Mater. Trans., 22A, 3019 (1990).
48. G.B. Schaffer and P.G. mcCormick, Scripta Metall, 23, 835 (1989).
49. G.B. Schaffer and P.G. McCormick, J. Mater. Sci. Lett., 9, 1014 (1990).
50. G.B. Schaffer and P.G. McCormick, Met. & Mater. Trans., 23A, 1285 (1992).
Questions
1. Explain the multi-layered structure of MA alloy-powder.
2. Explain the phenomenon of necking in brittle-brittle system dur-
ing MA.
3. How can microdeformation take place in a brittle-brittle system
during MA?
4. What are the various factors responsible for MA of a brittle-brittle
system?
5. Why was LQ model suitable to explain amorphization by MA?
6. Explain SSAR achieved by MA. How the MA amorphization in
systems with positive heat of enthalpy of mixing is caused.
7. Explain role of strain & plastic deformation in MA and MG.
8. Explain amorphization in intermetallic compounds.
Why is it not possible to have SSAR in a pure element?
9. Why are nanocrystalline grains considered to be defect free?
10. Explain the role of plastic strain during MA nanocrystallization.
11. Explain MA nanocrystallization in alloys.
12. Differentiate between critical temperature T crit and ignition tem-
perature T ig .
13. How MA helps in promulgating the solid state chemical reactions?
14. What is the effect of milling parameters on solid state chemical
reactions?
15. What condition must be satisfied for ignition to start during MA?
54
Energy Transfer and Energy Maps in Mechanical Alloying
E b = (1n b ) E 0 (6.1)
55
Mechanical Alloying
in a temperature rise of both the ball and the plate. A small fraction
of it is stored in the material as structural disorder. During MA, af-
ter a milling period of 10 min to 1 hr, depending on the materials, the
balls and chamber walls get coated with a thin powder-layer, and the
balls remain glued after collision. The collision behaviour approaches
that of the elastic type and the energy dissipation becomes minimum
(Fig.6.2).
In Fig.6.2, the impact energy (x-axis) can be regarded as the intensive
milling variable which can be correlated with the impact pressure and
temperature rise required to overcome the activation energy barrier. A
solid state reaction can occur only if the impact energy is high enough
to force the atoms climbing over the activation energy of that reac-
tion. The y-axis represents the quantity of energy released for a col-
lision at any given impact energy. This extensive variable can be cor-
related to the total quantity of energy, i.e. total time, needed in or-
der to complete a given reaction.
Various investigators have developed energy maps for MA/MG for
the intermetallic formation (Pd3Si [2], NiAl [3]) and amorphization (NiAl
[3], TiNi [4], NiZr [5]). These maps give information about the minimum
energy required to start the process and the total energy required for
the process to complete. Figure 6.3 illustrates the energy map for the
formation of nanocrystalline NiAl intermetallic by MA of Ni and Al el-
emental powders [3]. It is very clear from Fig.6.3 that a minimum total
energy E t = 200 kJ/kg is essential for initiation of the phase forma-
tion and 450 kJ/kg for the formation to be complete. It also points out
that the completion of the reaction between Ni and Al leading to the
formation of single phase NiAl is governed by E t rather than the en-
ergy transferred in each impact E b .
Fig.6.2 Energy dissipation during collision, E, vs kinetic energy of the ball E0 (Ref.1).
Fig.6.3 (right) Energy map of NiAl formation reaction during MA (Ref.3).
56
Energy Transfer and Energy Maps in Mechanical Alloying
References
1. M. Magini, Mat.Sci.Forum, 88-90, 121 (1992).
2. M. Magini and A. Iasomna, Mater. Trans. JIM, 36 123 (1995).
3. J. Joardar et al, In: Recent Advances in Metallurgical Processes, D.H. Sastry, et
al (eds), New Age International Publishers, New Delhi (1997), p.647.
4. B.S. Murty et al, Acta Metall., 43, 2443 (1995).
5. E. Gaffet, Mater. Sci. and Eng., A132 181 (1991).
Questions
1. What is the main mode of energy transfer in an attritor mill, a
vibratory mill and a planetary mill?
2. Describe phenomena of energy transfer during MA.
3. What are the energy maps used in MA?
4. What is the importance of energy maps in MA?
5. Draw energy maps for MA of (i) Al+Ni (ii) NiAl.
57
Mechanical Alloying
7 CONSOLIDATION OF
MECHANICALLY ALLOYED POWDERS
7.1 CONSOLIDATION TECHNIQUES
The MA powders have flake-like imperfectly packed layers and rough
surfaces which lower their flow rate due to extensive mechanical in-
terlocking. These particles are under heavily worked conditions, which
also lowers the shear induced bonding during cold compaction. Thus,
conventional methods of consolidation, cold compaction and sintering
are generally not suitable for their consolidation. Moreover, due to the
layered structure of these powder particles they have a high amount
of adsorbed and entrapped gases, which makes degassing, prior to con-
solidation, an essential step.
In the case of nanostructured and amorphous materials, the prob-
lems of consolidation are still severe. In the case of nanostructured
materials, there is rapid grain growth at elevated temperatures. Grain
size thermal stability in these materials can be given by a modified
Arrhenius equation [1]
58
Consolidation of Mechanically Alloyed Powders
59
Mechanical Alloying
60
Consolidation of Mechanically Alloyed Powders
61
Mechanical Alloying
where e& is the mean strain rate, Q is the activation energy of the rate-
controlling process in the deformation mechanism, R is the universal
gas constant and T is absolute temperature. Such strain rates lead to
increased substructural strengthening. For example, extruding at a rela-
tively low temperature and high speed produces a high value of z, which
can lead to a fine array of subgrains. Hot working at too high value
of z can result in sufficient stored energy to cause recrystallization.
Coarse elongated grains, e.g. in dispersion-strengthened superalloys,
are developed by secondary recrystallization of the materials.
Thermomechanical processing produces a high level of stored strain
energy [19]. It is a complex matter to control the processing history
of the material from the condition of milling, through the temperature,
ratio and speed of extrusion, to the details of post extrusion process-
ing, including hot or cold rolling or forging, and the grain coarsening
anneal. Achievement of the highest strength attainable by an MA
material demands the closest control of production parameters. The
recrystallized grains exhibit a texture ({100} <001> or {110} <001>)
and the grains are anisotropic (Fig.7.4). The texture is not only influ-
enced by the recrystallization temperature but also by the temperature
gradient conditions [2022]. Therefore, the desired recrystallized tex-
ture must be promoted by controlled nucleation and growth of these
very large grains during zone annealing, analogous to what occurs in
directional solidification. The development of anisotropic grains in these
alloys is attributed to the alignment of oxide particles along the extrusion
direction. The alignment is expected to be strongest along the surface
regions where the deformation imparted during extrusion is most intense.
62
Consolidation of Mechanically Alloyed Powders
References
1. J. Ravers et al, Met. and Mater. Trans., 27(A), 3126 (1996).
2. C.P. Dogan et al, Nanostruc.Mater., 4(6), 631 (1994).
3. D.G. Morris and M.A. Morris, In: Proc.Conf. New Materials by Mechanical
Alloying, E.Artz and L. Schultz (eds), Deutsche Geselschaft fr Metallkunde,
Germany (1989), p.143.
4. W. Smarsly and W. Bunk, Powd.Met.Int., 17, 63 (1985).
5. R.L. Anderson and J. Groza, MPR, 43 (4), 272 (1988).
6. J.A. Pordoe, Powd. Metall., 22, 22 (1979).
7. C. Suryanarayana, Nanostruc. Mater., 2(5), 527 (1993),
8. D.G. Morris, Mat. Res. Soc. Symp. Proc., 28, 145 (1984).
9. D.G. Morris, Metal Sci., 14 215 (1980).
10. M.A. Meyers et al, J. Metals, 33, 21 (1981).
11. R.N. Wright et al, Adv. Mater. and Proc., 10, 56 (1987).
12. O. Dominguez and J. Bigot, Nanostruc. Mater., 6 (8), 877 (1995).
13. G.E. Fougere et al, Nanostruc. Mater., 5 (2) 127 (1995).
14. K. Kobayashi, J. Jpn. Soc. Powd. Powd. Met., 42 (2), 191 (1995).
15. G. Frommeyer, Paper presented at Euro 95, Birmingham (1995).
16. Y. Hashimoya, J. Jpn. Soc. Powd. Metall., 42 (2), 185 (1995).
17. M.J. Fleetwood, Mat. Sci. and Techn., 2, 1176 (1986).
18. J.R. Pickens: ASM Metals Handbook, Tenth Edition, 2 200 (1990).
19. R.C. Benn et al, Powd. Metall., 24, 191 (1981).
20. K. Murakani et al, Met. Trans., 24A, 1049 (1993).
21. W. Sha and H.K.D.H. Bhadeshia, Met. and Mater. Trans., 25A, 705 (1994).
22. T.S. Chou and H.K.D.H. Bhadeshia, Met. Trans., 24A, 773 (1993).
63
Mechanical Alloying
64
Mechanical Properties of Mechanically Alloyed Materials
8 MECHANICAL PROPERTIES OF
MECHANICALLY ALLOYED
MATERIALS
Mechanical alloying enables the effective superimposition of numerous
strengthening factors including:
oxide dispersion
carbide dispersion
fine grain
high dislocation density and substructure
solid solution strengthening
Both direct and indirect effects influence the mechanical properties
of these materials. The interaction of the second phase particles with
dislocations is an example of a direct effect, while grain morphology
and texture are both indirect effects.
Moreover, the aforementioned five strengthening contributions can
be augmented by precipitation strengthening as well as intermetallic
dispersion strengthening. Thus, the resulting enhancement in mechanical
properties of MA materials is far greater than can be achieved by
conventional methods or say by the rule of mixture.
65
Mechanical Alloying
Fig.8.2 Grain size dependence of the yield strength of MA materials (The Int. J. Powd.
Met., 24(4), J.S.C. Wang et al, Microstructures and mechanical behaviour of mechanically
alloyed nickel aluminide (1988), p.315) (with permission from Prof. C.C. Koch)).
y = k m + k 0 d 1/2 (8.2)
0.13 Gb
s Or =
b
l ln rd / n g (8.3)
LF p I
l = MG J
0. 5
-2
OP F 2 I 0 .5
rd
MNH f K PQ H 3 K (8.4)
Hv
sy = km( m -2) (8.5)
(3.0 0.1)
67
Mechanical Alloying
Hv
sy = (8.6)
(3.0 0.1)
68
Mechanical Properties of Mechanically Alloyed Materials
8.2 FRACTURE
In general, the ductility and toughness of MA materials is found to be
dependent on the morphology of prior particle boundaries [11,12]. The
main strengthening factor is found to be the grain size. Elongation is
shown to be greater at dynamic strain rates, which is attributed to the
initiation of cracks.
Dispersoids can harden the matrix by obstructing the motion of dis-
locations. The dislocation may circumvent the obstacles by passing,
cross-clipping, climbing, etc [13,14]. The fracture appears to be domi-
nated by the disaccomodation between the dispersoid particles and the
matrix. The fracture path presumably follows crack initiation at the
dispersoid oxide/matrix interface and to the extent that dispersoids deco-
rate grain boundaries as well as reside in grain interiors. The fracture
paths may be transgranular or intergranular leading to a mixed mode.
In the case of nanograin materials, factors such as:
development of secondary cracks;
development of cracks within the nanograin particles and subse-
quent crack arrest at the particle nanograinparticles and subsequent
crack arrest at the particle nanograinmicrograin interface;
crack propagation along the boundary between the nanocrystalline
and submicron grain regions increases the fracture toughness [1].
8.3 CREEP
Studies on creep behaviour of various fully dense compacts of MA
dispersion-strengthened materials [11,12,15,16] and aluminides [17] have
revealed that MA enhances creep resistance and microstructural sta-
bility. These improvements are attributed to the fine scale dispersion
(~ 30 m) throughout the matrix, at matrix intermetallic interfaces, and
on subgrain boundaries which inhibit coarsening, recovery and
recrystallization.
The improvement in creep strength, even in the presence of stable
fine grain microstructures, in the MA materials may be attributed to
the ability of the dispersoids to interact with mobile dislocations over
a much wider range of temperature compared to the precipitates present
in the conventional alloys [14]. Diffusional creep is generally consid-
ered to increase with decreasing grain size. Artz et al [18] proposed
that when diffusion distances are small, elevated temperature deformation
processes are controlled by the motion of dislocation along grain bounda-
ries rather than the lattice diffusion vacancies. The creep rate is then
determined by the resistance offered to the motion of the boundary dis-
locations by dispersoids situated along grain boundaries in much the
same way as Orowan strengthening. The fracture path is presumably
dominated by fracture initiation at the dispersoid/matrix interfaces in
69
Mechanical Alloying
these materials.
The low creep rates in the case of coarse-grained dispersion-strength-
ened materials (e.g. thermomechanically treated MA superalloys) can
be explained by the modified power-law diffusional creep, which is usu-
ally applicable for high temperature alloys, as given below [14]:
e& = km D LM s u - sth OP n
(8.7)
N E Q
where e& is the creep rate (strain rate), u is the tensile strength, th
is the threshold stress below which creep rates are considerably smaller,
k m is the material constant, D is the diffusivity, E is Youngs modu-
lus and n is the stress exponent.
The threshold stress th, proposed in modified power-law creep,
Eq.(8.7), is assumed to depend on a parameter k r, describing the re-
laxation of the dislocation and is proportional to the Orowan stress,
References
1. J. Rawers et al, Met. and Mater. Trans., 27A, 3126 (1996).
2. T.G. Neil and J. Wadsworth, Scripta Met. and Mater., 25, 955 (1991).
3. K.Y. Wang et al, J. Mat. Sci. Lett., 12 (23), 1818 (1993).
4. W.E. Frazier and M.J. Koczak, In: Dispersion Strengthened Aluminium Alloys,
Y.W. Kim and W.M. Griffith (eds), TMS, Warrendale, PA (1988), p.573.
5. A.J. Heron and G.B. Schaffer, In: Advances in Powder Metallurgy and Particulate
Materials, V.6, A. Lawley and A. Swanson (compilers), MPIF and APMI, NJ
(1993), p.77.
6. S.K. Ganapathi and D.A. Rigney, Scripta Met. and Mater., 24, 1675 (1990).
7. R. Birringer et al, Defect and Diffusion Forum, 59, 17 (1988).
8. J.S.C. Jang and C.C. Koch, Scripta Metall. and Mater., 24, 1599 (1990).
9. J.R. Cahoon et al, Met. Trans., 2, 1979 (1971).
10. I.A. Hawk et al, Met. Trans., 19A, 2363 (1988).
11. C. Zakine, Creep mechanism in mechanically alloyed oxide dispersion strengthened
alloys, Presented at 1994 Powder Metallurgy World Conference, Paris, June 1994.
12. H.R. Last and R.K. Garrett, Met. and Mater. Trans., 27A (3), 737 (1996).
13. J.S.C. Wang et al, Inter. J. Powd. Met., 24 (4) 315 (1988).
70
Mechanical Properties of Mechanically Alloyed Materials
14. R.C. Benn and P.K. Mirchandani, In: New materials by mechanical alloying, E.
Arzt and L. Schultz (eds), Deutsche Geselschaft fr Metallkunde (1989), p.19.
15. E.L. Erich, Report-AFML-TR-79-4210, US Air Force (1980).
16. A. Lawley and M.J. Koczak, Report No. AFOSR-TR-86-0567 (Oct. 1985).
17. J.W. Pyun and S.I. Kwun, J. Korean Inst. Metals, 33 (6), 814 (1996).
18. E. Arzt et al, Acta Metall, 13, 1977 (1983).
Questions
1. What are the contributing strengthening factors in MA materials?
Comment on them.
2. How do tensile properties depend on the grain size in MA materi-
als?
3. How can the inverse Hall-Petch relationship be explained in the
case of MA intermetallic nanograin materials?
4. Comment on relationship between yield strength and hardness of
MA alloys.
5. Comment on fracture toughness of
(i) MA alloys;
(ii) Nanograin materials.
6. Explain mechanism of creep in
(i) MA fine-grained materials;
(ii) MA coarse grained (thermomechanically treated) materials.
7. Why is SCC resistance high in the MA alloys?
71
Mechanical Alloying
9 MODELLING MECHANICAL
ALLOYING
9.1 INTRODUCTION
Modelling can be simply defined as the representation of a physical
entity or a process by another physical or conceptual entity or a process
with a view to obtain greater insights into the behaviour of the sub-
ject. Models developed for complex processes cannot be expected to
be absolutely precise. Rather, they are intended to identify important
parameters, define the functional dependence of the process output on
process variables and predict results with an acceptable level of pre-
cision. One useful result of such process modelling is a considerable
reduction in the empirical studies needed to refine a process into a
useful engineering tool.
Some aspects regarding what is occurring during the MA process
have been known qualitatively for some time, but the description of
the MA process is complex and manifested as it involves concepts of
mechanics, mechanical behaviour, heat flow, thermodynamics and ki-
netics. In spite of this, modelling of MA has been an avenue of re-
cent interest for further understanding.
Though the models for MA are in the early stage of development,
various physical/theoretical models developed can be subdivided into
mechanistic, atomistic, thermodynamic and kinetic. Mechanistic models
deal with powder deformation, coalescence and fracturing, and ball dy-
namics in a milling device. Atomistic models permit a deeper insight
into the physics of non-equilibrium phase formation using the CALPHAD
method of thermodynamics data fitting and phase diagram calculations.
Kinetic models consider the competitive diffusion-controlled growth
kinetics of crystalline phase [1]. In the following section, these are
discussed in brief.
72
Modelling Mechanical Alloying
ditions. Mechanistic models can be classified into two types: local and
global:
1. Local modelling describes the various effects (thermal and me-
chanical) and events (deformation, fracture and welding) that transpire
when powder particles are entrapped between two colliding or sliding
surfaces. Thus, local modelling is generic in the sense that parameters
(relative impact velocity, angle of impact, charge ratio, etc.) affect-
ing the various events are common to all mills, although the values of
some of the parameters (e.g. relative impact velocity) are specific to
a particular type of mill and its operating conditions.
2. Global modelling is device specific. For example, this type of
modelling entails the study of factors such as distribution of impact
angles and heterogeneity of powder distribution within the mill, fac-
tors which clearly differ from one type of device to another. Several
efforts have been made at analysing the ductileductile system (to start
with) and these are described here.
Fig.9.1 Various kinds of fragmentation and coalescence events that can be imagined
to occur during MA depending upon impact angle (Ref.4, Met. Trans., 21A, 289 (1990)).
73
Mechanical Alloying
of the total number of impacts [3], only the normal collisions are con-
sidered.
The structure developed in the MA powder depends on the defor-
mation, coalescence and fracture events taking place during the process.
Therefore, a model for the deformation, coalescence and fracture of
MA events has been developed under the following assumptions [4]:
only the ballpowderball collisions are considered as they are
greatest in number [2];
during the processing, a powder coating of about 100 m thickness
gets [2] coated uniformly;
it is assumed that the powder particle shape is a oblong spheroid,
though it varies from spherical to flake. This shape is characterised by
a ratio (shape factor) of minor and major axis of the spheroid;
it is assumed that these oblong spheroid shaped particles rest on
a grinding ball with their major axis parallel to the ball surface.
Deformation
During the collision, the kinetic energy of the balls is converted into
deformation energy during the approach of their centres. In the case
of alloying of two or more phase materials, it is taken as the resist-
ance offered by the softer component present. The stress homologous
to this energy conversion is applied on the deforming particles. Assuming
that the bulk powder particles undergoing collision constitute an in-
dividual particle, the deformation may be expressed as a function within
the contact area as
F r IJ
a (r ) = R v G
1/ 2
-
r2
(9.1)
HH K
b 0
V Rb
e = - ln
LM h - a (r) OP
0
(9.2)
N h Q 0
74
Modelling Mechanical Alloying
Coalescence
During plastic deformation, the brittle oxide layer present on the particle
surface breaks, the particles flatten in compression, their surface area
increases and the underlying metal is progressively exposed. The bonding
force F w, acting at the weld that is formed, can be given as
F w = A w u (9.3)
where u is the tensile strength of the weld and Aw is the effective new
surface area created. If the same metal welds together, u is the tensile
strength of the bulk material. When a cold weld is made between two
lamella particles, the weld strength is given by the rule of mixture.
Using the relationship, H v = 3 u, cf. Eqs. (8.6) and (9.2), one can
calculate the bonding force (F w) acting at the weld point which must
be more than this for fracture to take place.
Fracture
Considering only the forging fracture which may develop over several
impacts (chances of shear fracture and dynamic fracture are much less
during the MA process), the cracks formed will grow radially along
the major axes of the particles (Fig.9.2). A crack initiates when the
critical tensile strain is attained over a critical length and the crack
propagation occurs when the plastic energy release exceeds a value
characteristic of the material. Thus, if the particle is sufficiently small
so that this length is greater than the particle size, the particle is con-
sidered below its comminution limit and will not fracture.
The following equation can be used to predict the tensile true fracture
strain f of a particle
z (r) = 2 f (9.4)
75
Mechanical Alloying
J = bsu e a p
F 3s IJ
mG
m +1
(9.5)
H 2s K u
ac = JIc
Km FG 2 IJ m+1
(9.6)
p m H 3 su K
Values of J Ic and ac are calculated based on the input values of the
material parameters K Ic, E, K, u and n, where E is Youngs modu-
lus of the powder material.
Using the expression for strain as a function of radial position within
the powder particle permits determination of the total approach between
balls needed in order to exceed some critical strain over a given length.
The condition for forging fracture can now be expressed as
a (r ) F a 2 f 2/ 3 I 0 .5
h0 GH
= 1 - 1 - c s2
4 Rw
JK exp( - e c ) (9.7)
FG
fsf = 1 -
a (r ) IJ 1.5
fsi (9.8)
H h0 K
where f sf and f si are the final and initial shape factors, respectively.
76
Modelling Mechanical Alloying
tc
dn ( v )
dt z
= - a f n ( v ) - a w n( v ) f (v' ) dv' +
0
+ 2a f z 1
v v'
a
z
2 0
v
m ( v' )dv' + w n ( v' ) f ( v - v' )dv'
(9.9)
n- J
-a w NJ fI (9.10)
I =1
77
Mechanical Alloying
Equation (9.10) does not allow particles to weld if the sum of their
volumes is larger than a maximum size class. The corresponding
discretised form of the third and fourth term on the right-hand side of
Eq.(9.9) are
n
NI
+2 a f I -1 (9.11)
I = J +1
J -1
aw
+
2
N J - I fI (9.12)
I =1
There are two special cases that must be considered during the
numerical analysis. The first deals with the smallest size class considered
(J = 1). Particles in this size class are not allowed to fracture. In ad-
dition, there are no smaller particles that could weld. The equation
applicable to this size class is thus written as
LMt dNJ OP = 2a f
n
NI
- aw
n- J
N J fI
N c
dt Q J =1 I = J +1
I -1 I =1
(9.13)
The second is for the largest size class (J = n). Particles of this size
range are not allowed to weld to any other particles. The equation for
this size class is
LMt dN J OP = -a f N J +
aw J -1
N J - I fI
N c
dt Q J =n 2 I =1
(9.14)
Fig.9.3 Cu, Nb and Cr particle size distributions obtained from model predictions are
compared to those found experimentally for 4 hrs milling (Ref. 8).
78
Modelling Mechanical Alloying
79
Mechanical Alloying
forging process between the two parallel plates [9]. We consider a po-
rous metal cylinder, composed of a large number of individual pow-
der particle, of height h 0 and radius r h, as shown in Fig.9.5. The cyl-
inder is impacted between colliding workpieces having an average relative
collision velocity v 0. During an impact event, it is assumed that par-
ticles confined in the cylinder are subjected to a true deformation strain
. It is also assumed that the rate of this deformation decreases lin-
early during the impact. That is, this velocity is v at the initiation of
compaction and is zero at the time () at which the impacting
workpieces begin to rebound from each other.
This model can be used, among other purposes, for estimating the
time required for MA. For this first select a unit cell or the milling
device, i.e. in a SPEX mill, then the unit cell is taken as the mill vol-
ume divided by the number of balls within it. In the attritor, the unit
cell volume is the average volume belonging to each ball.
For such a unit cell, the powder volume strain per unit time is
r h2 h 0/t c, where is the deformation strain per impact and tc is the
time between impact events. To alloy the material, the critical defor-
mation strain c, Eq.(9.4), must be generated throughout the total pow-
der volume v p. The processing time t p is the product of V p and c di-
vided by the volume strain total per unit time [10]
t p = (V p c t c)/(r h2 h 0 ) (9.15)
t c = /v 0 (9.16)
where is the mean free path between collisions. The velocity of balls
in an attritor mill can be estimated by considering them as a fluid.
As the agitation speed increases, the balls rise higher against the chamber
wall (Fig.9.6). From elementary fluid mechanics, the height of a fluid
above the reference line is given by [10,11]:
80
Modelling Mechanical Alloying
h f = 2 R 2 /4g (9.17)
2 0. 5
w=
R
d i
g hf (9.18)
81
Mechanical Alloying
2x
v0 = (9.19)
t
where x is the vial length. For a typical mill, the average velocity
is found to be 3.9 m/sec. These velocities are in good agreement with
the studies of Davis et al vis-a-vis a combination of computer simu-
lation and experiment [3].
The impact velocity in the case of a planetary mill can be calcu-
lated using the following relationship [12]
LM w F R - R I
3 b OP
v0 =M
H 2 K +w
v v
p wv
F RI
R P R - b
MM w p
p
PP H 2 K
v
(9.20)
N Q
where v and p are the angular velocity of the vial and the plate, re-
spectively, and R v and R b are the radius of the vial and the balls, re-
spectively. The impact velocity v 0 in planetary ball mills for typical
conditions is 11.14 m/sec [13]. This impact velocity is an order of mag-
nitude higher than the impact velocity in an attritor and is a primary
reason why processing times in a Spex mill or planetary mill are less
than in other MA devices. Estimated collision velocities, times between
impacts and mean free paths for common MA devices are given in Table
9.1.
For horizontal mills, the diameter of the drum varies from 1 to 2
m and the impact velocity ranges from 4.4 to 6.3 m/sec. These are
fairly high. The long processing times required by these mills are re-
lated to the longer time between collisions and/or the amount of ma-
terial impacted per collision.
Table 9.1 Typical impact velocity and impact frequency for common MA devices
82
Modelling Mechanical Alloying
b g
2 t = gt v0-0.2 r / E
0. 4
Rb (9.22)
where Rb is the ball radius, is the ball density, E is the tensile modulus
of the colliding media and gt is a parameter which depends on the col-
lision geometry, i.e. ball on ball, ball on plate (value is an order of
ten). If the ball and colliding surfaces are of different materials, Eeff
will be taken.
83
Mechanical Alloying
rh = gr v00.4 F rI 0. 2
Rb
H EK (9.23)
Pmax = g p v00.4 F rI 0. 2
E
H EK (9.24)
84
Modelling Mechanical Alloying
z
Vp s u de = Vp Cp DT (9.25)
85
Mechanical Alloying
86
Modelling Mechanical Alloying
-Vp Cp
F dT I = h AbT - T g
H dt K t a (9.26)
t = -8.53 10 -3 Cp
F rp2I ln bT - T g
a
GH k JK bT - T g
i a
(9.27)
FG R IJ - FG R IJ
i
a
i
c
h =
HR K HR K
l l
i
FG R IJi
c
(9.28)
HRK l
87
Mechanical Alloying
Table 9.3 Relative changes in bond lengths for fcc, bcc and hcp lattices during
amorphization
Lattice 1 2 3
vA 1
CBmin = l 0 = l0
vB - vB FG r IJ
B
3
-1 (9.29)
Hr K
A
88
Modelling Mechanical Alloying
where 0 is the atomic size factor, v A and v B are the atomic volumes,
respectively.
Such models have been successful in predicting the amorphization
even in the system with +H mix , as discussed in Section 5.2.1.
References
1. B.B. Khina and F.H. Froes, J. Metals, 7, 36 (1996).
2. R.W. Rydin et al, Met. Trans., 24A, 175 (1993).
3. R.M. Devis et al, Met. Trans., 19A, 2867 (1988).
4. D.R. Maurice and T.H. Courtney, Met.Trans., 21A, 289 (1990).
5. B.J.M. Aikin et al, Mat.Sci.and Engn., A147, 229 (1991).
6. J.W. Hilgers et al, Math.Mod., 6, 463 (1985).
7. K.L. Kuttler et al, Appl. Analysis, 19, 75 (1985).
8. B.J.M. Aikin and T.H. Courtney, Met.Trans., 24A, 2465 (1993).
9. D.R. Maurice and T.H. Courtney, Met.Trans., 21A, (2) 289 (1990).
10. T.H. Courtney and D.R. Maurice, In: Solid State Powder Processing, A.H. Clauer
and J.J. deBarbadillo (eds), The Minerals, Metals and Materials Society, Warrendale,
PA (1990), p.3.
11. V.L. Strater, In: Fluid mechanics, McGraw-Hill, New York (1951), p.25.
12. M. Magini, Mat.Sci.Forum, 88-90, 121 (1992).
13. M. Abdellaoui and E. Gaffet, Acta. Metall. and Mater., 43 (3), 1087 (1995).
14. D. Maurice and T.H. Courtney, Met. and Mater. Trans. A., 26A, 2431 (1995).
15. D. Maurice and T.H. Courtney, Met. and Mater. Trans. A., 26A, 2437 (1995).
16. S.C. Lim and M.F. Ashby, Acta. Metall., 35, 1 (1987).
17. Y. Chakk et al, Acta. Metall., 42 (11), 3676 (1994).
18. P.L. Brun et al, Scripta. Met. and Mater., 26, 1743 (1992).
Questions
1. With the help of a schematic show various possible fracture & coa-
lescence events which may take place during MA.
2. Give a typical account of impact velocity and impact frequency in
different MA devices; which mill is considered to be most effec-
tive?
3. Describe how the volume of powder impacted in single collision
can be estimated.
89
Mechanical Alloying
90
Joining of Mechanical Alloying Materials
10 JOINING OF MECHANICAL
ALLOYING MATERIALS
Despite the degassing of MA powders prior to their consolidation, MA
products still contain gaseous components which impair the joining of
these materials. Various potential joining techniques applicable to MA
materials are described here.
10.1 WELDING
In general, fusion welding is not suitable for joining MA materials. This
technique causes the dispersoid particles to be rejected from the molten
metal and the development of a high level of porosity in the weld zones
due to excessive degassing (Fig.10.1) [1]. Nevertheless, sound TIG
welds suitable for positioning or non load-bearing joints can be made.
Fusion welding processes, which minimize the size of the molten zone,
produce relatively sound welds. Spot and resistance seam welds with
excellent tensile strength can be made. However, it may never be pos-
sible to achieve the full stress rupture properties of the base material.
Solid state welding techniques, friction welding, diffusion welding,
explosive welding and magnetostrictive welding have demonstrated sound
weld joints [2]. Figure 10.2 shows the microstructure of a Dispal/Dispal
friction weld made by using a continuous drive type friction welding
1 mm
Fig.10.1 Optical micrograph of inert gas tungsten-arc melted Dispal showing porosity
in fusion region. (Ref.1, W.A. Kaysser and W.J. Huppmann (editors)), Horizons of Powder
Metallurgy, Part II, (1986), p.710 (Verlag-Schmid GmbH)).
91
Mechanical Alloying
500 m
30 m 0.9 m
Fig.10.2 Microstructure of Dispal/Dispal friction weld; (a) and (b) optical micrograph;
(c) TEM micrograph of the centre of the friction weld zone (Ref.1).
Fig.10.3 Continuous drive friction welding: basic principle (a); process variables (b).
machine [1]. During the process, one of the components is held sta-
tionary, while the other one is rotated at a constant speed (Fig.10.3).
As the two surfaces continue to rub against each other, heat is gen-
erated at the interface. After a certain weld-time, the rotating component
is stopped rapidly and the axial pressure is increased. Thus, the hot
material cools under pressure and the weld is consolidated. Due to
cooling of the molten pool under pressure, the welds have a fine micro-
structure without porosity (Fig.10.2), which leads to sound welds. The
variation of hardness within the weld zone is shown in Fig.10.4. Hard-
ness in the weld zone remains constant in the case of Dispal/Dispal,
but increases in the case of Dispal/steels. The bond strength in the
welds is given in Table 10.1.
Diffusion welding is performed by giving a suitable heat treatment
to the well-aligned polished surfaces. Diffusion welding of polished MA
92
Joining of Mechanical Alloying Materials
6000 (see Section 12.1) surfaces have been performed by giving a heat
treatment at 1180C for 2 hrs [1]. The welded joint produced had good
tensile properties. However, the stress rupture properties in the bond
area were inferior to the base material due to fine grains. The prob-
lem of fine grains can be overcome by special surface treatment rec-
ommended by Helko [3]. However, the welding cycle used by Helko
is time consuming and requires well-aligned parallel surfaces as well
as a suitable atmosphere to get a sound weld everywhere.
10.2 BRAZING
Brazing is applied mainly for attachment of MA components in the
aerospace industry. The joint strength is of course limited by the strength
of available braze filler materials [2].
93
Mechanical Alloying
References
1. H. Kreve et al, In: Horizons of Powder Metallurgy, W.A. Kaysser and W.J.
Huppmann (eds), Verlag Schmid GmbH (1986), p.707.
2. J.J. Fisher and J.H. Weber, In: Proc. of Conf. on Structural Application of Mechanical
Alloying, F.H. Froes and J.J. deBarbadillo (eds), ASM International, Materials Park,
OH (1990).
3. K.H. Helko, US Patent 3, 787, 748 (1974).
4. R.F. Singer and G.H. Gessinger, Met. Trans., 13A, 1463 (1982).
5. H. Rydstad and R.F. Singer, Ibid Ref.1, p.713.
Questions
1. Why fusion welding techniques are not suitable for MA materials.
2. Give one example where diffusion welding has been used success-
fully for MA materials.
94
Joining of Mechanical Alloying Materials
95
Mechanical Alloying
96
Rapid Solidification and Mechanical Alloying
1. REFINEMENT OF MICROSTRUCTURES
High strength alloys can be developed through compositional modification and suitable
heat treatments. refined microstructures and higher solute contents, in conjunction with a
uniform distribution of fine precipitates during subsequent ageing of the RS alloy, have
been found useful for this purpose. For example, addition of about 1% of Fe, Ni, Zr or
Mn to conventional 7xxx series aluminium alloys improve their mechanical properties.
RS of these alloys results in about 20% increase in YS and UTS without loss of ductility
(2). These alloys also possess improved corrosion resistance and stress corrosion
cracking resistance. Similarly, addition of eutectoid formers (e.g. Co,Cr, Cu, Fe, Ni etc.)
to titanium using RS leads to much higher strengths than normally possible through IM
techniques. Table 11.2 lists the room temperature mechanical properties of RS alloys
showing the significant improvement achieved through this route. A similar but better
improvement in the properties of MA Alloys can be achieved using MA techniques
(Table 113).
3. DISPERSION STRENGTHENING
One of the serious difficulties encountered in the precipitation hardened alloys is the
coarsening of the precipate on elevated temperature exposure and consequent degradation
of the mechanical properties. Thus, development of high-temperature alloys requires very
low equilibrium solid solubility of the solute elements and a very low diffusion rate.
Addition of Fe, Cr and V transition metals such as to aluminium (4,5), and of rare-earth
elements like Er and Nd to titanium (6) by RS techniques are found to be useful in
retaining the reasonably high temperature strength levels in these alloys. The transition
elements form fine, thermically stable intermetallic dispersoid particles whereas the
rare-earth elements form uniformly dispersed oxides in situ. However, sometimes regions
near the grain boundaries are devoid in these materials (Fig.11.1).
It may be emphasized that MA is a much more powerful technique for dispersion
strengthening. Because of the heavy working involved, the matrix grain size is very fine,
occasionally reaching the nanometer levels, and further the distribution of dispersoids is
much finer and uniform throughout the matrix, at the interface, and on the subgrain
boundaries. MA enhances microstructural stability at elevated temperatures (Fig.11.2) ,
is attributed to the presence of fine scale dispersoids in the microstructure which inhibit
coarsening, recovery and recrystallization (7).
97
Mechanical Alloying
conventional methods is difficult becuase magnesium boils before titanium melts; but MA
allow about 3% Mg dissolution in titanium (10). further, MA can produce homogeneous
solid solutions in liquid-immiscible (CuPb) and solidimmiscible (CuFe) systems
(11).
5. AMORPHIZATION
The free-energy diagrams perdict that the homogeneity range of amorphous alloys formed
by RS is usually divided into relatively narrow regimes located near deep eutectics in
the liquidus (refer Fig.5.8).
In contrast, the composition range of the single phase amorphous alloy prepared by solid-
state reaction is wide and continuous, and is located near the centre of the composition
range. (Fig.11.3) (12). Mechanical alloying prevents intermetallic phase formation during
interdiffusion because of extremely thin layer, thus resulting in a wider glass forming
range(13).
6. MA amorphous phases have a relatively more relaxed disordered structure with a more
developed short range ordering and contains a large milling induced stored energy.(14)
7. Amorphization in many systems like Ni-Pd, and Ni-Ti intermetallic is not possible by RS
technique.
But MA made the amorphization possible in such cases.(12)
The MA amorphouse alloys and rapidly quenched alloys, both exhibit very similar
properties like crystallization temperatures and atomic structures.
1M 7075
T6 503 572 10
RS
Extruded 634 717 9
7075(+1/Ni+1/Fe)
T7 Extruded 580 614 12 47
RS CW67
T6 595 645 3.5
MA 7075
848 986 13
1M Ti6A/4V
1090 1110 2.5
RS Ti6A/4V+1B
T4 277 464 22 32
1M 2024
326 542 24
RS 2024
T4 500 570 12 30
MA IN 9021
T6 560 600 12 44
Aluminium system
Rapidly solidified plus mechanically alloyed (RSMA) aluminium
titanium alloys have been shown to have elevated temperature mechani-
98
Rapid Solidification and Mechanical Alloying
Fig.11.1 TEM micrographs showing the size and distribution of dispersoids in: a) RS
Ti3Al2% Er alloy;(b) MA Ti3Al20%Nb2% Mo3%V2% Er (Ref.1, Light Metal Age,
6, 18 (1989)).
Fig.11.2 Grain growth in heat treated ribbon, extruded ribbon-powder, and extruded
and heat treated MA powder of Cu5% Cr (Ref.7, E.Artz and L. Schultz (editors)),
New Materials by Mechanical Alloying (1989), p.3 (Deutsche Gesellschaft fr
Metallkunde).
99
Mechanical Alloying
100
Rapid Solidification and Mechanical Alloying
MSMA 185 alloy has an ambient temperature yield strength (526 MPa)
106% greater than that of the RS Al3%Ti3% Ce alloy (255 MPa).
At 300C, the yield strength of MSMA 185 alloy (172 MPa) is 25%
greater than that of MS Al3%Ti3%Ce alloy (137 MPa). Elongation
in the MS alloys is found to be 22% (Fig.11.5). The MA, however,
reduces ductility. AM6 alloy has an elongation of 9% vis-a-vis alloy
AT6 of 22%. Similarly, MA Al3%Ti3% Ce has a tensile elongation
of 5%, but the MS Al3%Ti3% Ce alloy has elongation of 21%. Os-
tensibly, the reduction in ductility can be attributed to the introduction
of oxide and carbide dispersoids during MA. These dispersoids pref-
erentially locate themselves on the grain boundaries and inhibit defor-
mation and limit plastic accommodation.
Ternary RS AlFeCe alloys can be used up to 327C. However,
these alloys lose strength rapidly at high temperatures (see Fig.12.12).
To overcome this deficiency, the RS Al8.4%Fe3.4%Ce alloy has been
mechanically alloyed in a Spex mill to uniformly disperse carbides,
oxides and intermetallic phases (metastable Al 10 Fe 2 Ce, and stable
Al13Fe3Ce and Al13Fe4 intermetallic crystalline) in the aluminium, which
results in the increase in strength and stiffness [16,17].
In another investigation using the RSMA process for Al4%Cu
1%Mg1.5%Fe0.75%Ce alloy, the strength level could be increased
to 435 MPa [18].
Titanium system
Two advanced titanium-based alloys, Ti1% Al8%V5% Fe1% Er and
Ti24%V10%Cr5% Er, both with extensive rare earth additions have
been developed using the RSMA process. The Ti1%Al8%V5%Fe
1% Er alloy consolidated directly from a gas atomised powder had
coarser beta grains of about 30 m. A MA of up to 40 hours [19] did
not yield a totally uniform structure, but the majority of the structure
in the consolidated alloy consisted of submicron grains with dispersion
of 30 to 50 nm in size. The fine-grain structure was found to be stable
with little grain coarsening, and a little drop in hardness was observed
on solution treatment at 675C.
The Ti24%V10%Cr5%Er alloy in the as-atomised condition was
badly segregated with large chunks of free Er. The microstructure of
the consolidated RSMA alloy consisted of both very fine grained re-
gions and some coarser grains, with dispersion in fine grained regions
of less than 10 nm in size. The RSMA alloy had higher hardness (437
in comparison with 304 KHN) due to the grain refinement and disper-
soids.
Thus, fine dispersions and stable grain refinements are possible by
101
Mechanical Alloying
References
1. C. Suryanarayana and F.H. Froes, Light Metal Age, 6, 18 (1989).
2. P.K. Domalavage et al, Met. Trans., 14, 1599 (1983).
3. W.E. Quist et al, In: Aluminium-Lithium Alloys, C. Baker et al (eds), The Institution
of London (1986), p.625.
4. P.S. Gilman et al, Industrial Heating, 56 (2), 30 (1989).
5. Y.M. Kim, In: Proc. Conf. on Dispersion Strengthened Aluminium Alloys, Y.M.
Kim and W.M. Griffith (eds), The Minerals, Metals and Materials Society,
Warrendale, PA (1988), p.157.
6. S.M.L. Sastry et al, Met. Trans., A15, 1451 (1984).
7. D.G. Morris and M.A. Morris, In: Prof. Conf. on New Materials by Mechanical
Alloying, E. Artz and L. Schultz (eds), Deutsche Geselschaft fr Metallkunde,
(1989), p.3.
8. C.R. Clark et al, In: Proc. Int. Conf. PM 2 Tech 95, M.A. Phillips and J. Porter
(eds), MPIF, Princeton (1995).
9. Z.A. Zhang et al, Met. and Mater. Trans., 25A, 73 (1994).
10. R. Sundresan and F.H. Froes, Key Eng. Mater., 29-31, 199 (1989).
11. R. Sundresan and F.H. Froes, J. Metals, 39 (8), 22 (1987).
12. E. Hellstern et al, J. Less-common Metals, 140, 1 (1988).
13. J. Eckert and L. Schultz, Mat. Sci. and Engn., A134 1389 (1991).
14. A. Inoue et al, Mat. Trans. JIM, 32 (2) 148 (1990).
15. W.E. Frazier and J. Cook, In: Solid State Powder Processing, A.H. Clauer and
J.J. deBarbadillo (eds), The Minerals, Metals and Materials Society, Warrendale,
PA (1990), p.257.
16. M.L. Ovecoglu and W.D. Nix, In: New Materials by Mechanical Alloying Techniques,
E. Artz and L. Schultz (eds), Deutsche Gesellschaft fur Metalkunde, Oberusel,
Germany (1989), p.287.
17. M.L. Ovecoglu et al, Met. and Mat. Trans., 27A, 1033 (1996).
18. P.S. Gilman and K.K. Sankaran, Ibid Ref.5.
19. R. Sundares and and F.H. Froes, 6th World Conference on Titanium, Cannes, France
(1988).
Questions
1. In what way are the two techniques MA and RS similar?
2. In what do mechanically alloyed materials differ from rapidly solidi-
fied ones?
3. Give two examples of amorphization reaction which can be achieved
by MA but not by RS.
4. Explain how magnesium can be added to titanium?
5. How AlLi alloys produced by RS techniques from from that of MA.
6. What advantages are associated with MA of rapidly solidified mate-
rials? Explain the phenomenon with the help of suitable examples.
102
Applications
12 APPLICATIONS
Since its advent, the MA technique has been applied to develop various
novel compositions and improve the performance of existing materials.
Many of these are produced at industrial level and find applications in
the commercial sector, while many are potential candidates to find an
application. This is discussed in the following section.
103
Mechanical Alloying
(particle size 150 m) is used. Yttrium oxide is introduced into the mix-
ture in the form of 1 m aggregates, each consisting of numerous par-
ticles of 20 to 40 nm diameter. The MA powder is then degassed in
vacuum at 538C and sealed in mild steel cans. These cans are then
extruded using an extrusion press at temperatures in the range 1010
1066C and at an extrusion ratio of 13:1 (Fig.12.1). The microstruc-
ture at this stage consists of incredibly fine (0.4 m) equiaxed grains
resulting from dynamic recrystallization during the hot deformation. Hot
rolling is carried out at 1025C with 20% thickness reductions per pass
with a total thickness reduction ranging up to 90% [2]. Afterwards,
zone annealing is carried out at 12301290C, which increases the grain
aspect ratio to more than 15 by a process of secondary recrystallization,
giving a structure reminiscent of directional solidification [3] and hence
the high-temperature strength of the alloy. In general, 4060% thick-
ness reductions are required to produce excellent directional
recrystallization. The recrystallized grains tend to assume plate-shaped
morphologies (see Fig.7.4).
Fig.12.2 TEM micrograph of recrystallized MA 760 superalloy, cooling rate 10C per
min (courtesy Dr H.K.D.H. Bhadeshia).
104
Applications
Alloy
Property
MA 754 MA 758 MA 6000 MA 760
105
Mechanical Alloying
Fig.12.3 Stress for 1000 hours life to rupture as a function of temperature for alloys
MA 6000, and other reference material.
12.2 MA STEELS
Many MA materials are now commercially available, but steels produced
using MA show particular promise in a variety of applications. For
example, MA 956 (Fe20% Cr4.5% Al0.5% Ti0.5% Y2O3) is a chro-
mium-rich, ferritic stainless steel containing aluminium for oxidation re-
sistance, (aluminium forms an Al 2O 3 protective layer at high tempera-
tures) together with a dispersion of Y 2O 3 particles for creep resistance.
An alternative ferritic stainless steel variant MA 957 (Fe13.5%Cr
0.3%Mo1%Ti0.3%Y2O3) contains titanium rather than aluminium, and
is designed for application in the nuclear industry. MA 956 can be used
at operating temperatures of over 1300C in corrosive atmospheres. The
107
Mechanical Alloying
b
Fig.12.6 Components made of MA 956: a) burner nozzles machined from forgings; b)
air-stream swirlers fabricated from sheet.
alloy is used as sheet material for aircraft and industrial gas turbine
combustors, in burners, swirlers and heat exchangers of power generation
equipment (Fig.12.6), and in heat treatment equipment (Fig.12.7) [5,1].
108
Applications
ODM 331 13 3
ODM 361 13 6
ODM 031 20 3
ODM 061 20 6
109
Mechanical Alloying
751 32 47 134
FeCrAl 35
F10R 165 16
able even at temperatures above 1100C, which is the upper limit for
the use of superalloys for both the mechanical properties and the oxi-
dation resistance. A suitable heat treatment of ODM has also been
developed to favour the transgranular rupture in these materials, which
improves ductility. When compared with MA 956, the time to rupture
is multiplied by a factor of 100. The superior mechanical properties
of these ODMs above 1000C is due to the strengthening effect of
dispersoids. Thus, it is envisaged that these ODMs may find applications
in energy conversion systems (particularly of heat exchanger tubes) as
well as other high temperature applications in corrosive atmospheres.
For the requirements of modern combustion engines, high wear re-
sistance steel having a tough matrix and hard, fine dispersoids is re-
quired. In this respect, four hard phases can be used, namely: NbC,
TiC, TiN and Al 2O 3 . All other hard phases are either too soft or un-
stable within the steel matrix at sintering temperatures. Sintered steels
containing hard phase in the range of (1025 vol.%) have been pro-
duced by MA of elemental composition in an attritor mill followed by
liquid phase sintering at 12801320C in a hydrogen atmosphere. A high
phosphorous content leads to a harmful embrittlement of the grain
boundaries, as a consequence the phosphorous content is chosen very
carefully [8,9]. The influence of this hard phase on the mechanical
properties of the composite material is given in Table 12.5. It can be
observed that the effect of wettabilities of different hard phases with
the matrix on sintered densities of the composite material is much less.
However, stiffness of the composite increases as a function of Youngs
modulus of the reinforcing component. The strength of Al2O3-containing
110
Applications
Table 12.5 Properties of the used hard phases and the composite materials with a
composition of Fe10 vol% hard phase 0.6% P0.9% C
material is 25% less compared to that of NbC, and the elongation frac-
ture is 50% less. Surprisingly, sintering densities are found to be in-
creased with the amount of an inert second phase in the composite
which does not take part in the sintering process. It has been attrib-
uted to the fact that sintering in the composite materials takes place
by means of intergranular diffusion for a longer period of time while
the grain size remains more or less stable. The wear behaviour of these
materials has been found to be the best when the hard phase content
is about 10 vol.%. It is also reported that dry wear is reduced using
coarser hard phase. For oil lubrication at higher surface pressures (800
N/mm 2) and longer times (100 hrs), finer hard phase particles lead to
less wear. Thus, for oil lubricated tribosystems, these hard phase con-
taining sintered steels offer a better alternative to conventional wear
resistance PM steels like sintered and forged T15 MSS. Moreover, the
matrix hardness can be controlled within a wide range by heat treat-
ment and, hence, can be matched to the respective application.
111
Mechanical Alloying
112
Applications
Low-density alloys
For the aerospace industry, an alloy with properties equivalent to or
better than those of the widely used 7075-T73 alloy and with lower
density is desired. For density reduction in aluminium, there are generally
only two choices of alloying additions that merit practical considera-
tion lithium and magnesium. The proportion of these elements which
can be tolerated is limited by the possible occurrence of stress-corrosion
113
Mechanical Alloying
Fig.12.11 Flow sheet of the production process for AlLiMg alloy (Ref.14, Powd.
Met. Int., 22(3), 22 (1990) (Verlag Schmid)).
115
Mechanical Alloying
the low density of the lithium, the lightest existing metallic element.
Based on the work of S.J. Domachie et al [15,16], IncoMAP has
developed the IN-905 XL alloy (Al4% Mg1.5% Li1.2% C0.4%O)
which matches 7075-T73 in strength but is 8% less dense and 15% more
stiff. The outstanding feature of the alloy is the minimal degradation
of properties when stressed in the transverse direction. This charac-
teristic is of great importance in forgings which are uniaxially stressed.
Its general corrosion resistance is about 100 fold better than that of
7075-T73. Heating cycles up to 450C do not alter the strength. This
characteristic can be important in service as well in manufacturing.
Figure 12.13 shows a prototype aircraft wheel forged of this alloy.
High-temperature alloys
The development of MA high-temperature aluminium alloy can be put
in three classes:
DISPAL
AlTi alloys
AlFeCe alloys
The MA process has lead to the development of Dispersion Strength-
ened Aluminium (DISPAL) high-temperature aluminium alloys [17]. The
alloy powders are processed directly by reaction milling with carbon
black and controlled additions of oxygen from the milling atmosphere
and with this independent control of the dispersoid content, the alloys
can suitably be tailored for applications. Already, forged automotive pis-
tons, high-temperature electrical conductors, interferometers and structural
parts in aerospace cryogenic fields have been made from various
DISPAL alloys.
116
Applications
Alloy Al Mg Cu Li C O
118
Applications
RS7075 (+1%Ni+1%Fe)
634 717 9
595 645 3.5
MA 7075
Dispal
(Al12%SiCuMgNi) 340 370 3 81 2.70 120
Fig.12.15 Comparison of creep rate for non-MA (RS) and MA Al8% Fe4% Ce alloy
(Ref.20).
119
Mechanical Alloying
120
Applications
124
Applications
have a binding energy of Ni 2p 3/2 less by 0.5 eV, which is the rate-
determining step for dissociative adsorption of hydrogen on nickel atoms.
The formation of Mg 2Ni is found to take place in the hydriding reaction
of the MA alloy, and practically all the nickel converts into the
intermetallic compound within 23 cycles. Mg 2 Ni is shown to form
mainly by
2MgH 2 + Ni Mg 2Ni + 2H 2
126
Applications
SmFe magnets
Magnetic alloy systems like Sm (Fe,X)12 and (Sm,Zr)Fe3 have also been
prepared successfully using an MA technique [38,39]. The MA of el-
emental samarium and iron powders leads to a two-phase mixture of
amorphous SmFe and -Fe. Heat treatment at 650800C for sev-
eral hours produces Sm 2 Fe 17, with a Th 2Zn17 crystal structure, which
has a Curie temperature of only 116C and an in-plane anisotropy. In
this state, the material is still soft magnetic because of the in-plane
anisotropy of Sm2Fe17. During the subsequent nitriding treatment at 400-
550C in a N 2 atmosphere, nitrogen atoms are introduced into its
Th 2Zn 17 crystal structure, resulting in an extension of the lattice pa-
rameter as calculated by the X-ray diffraction patterns (Fig.12.23b and
127
Mechanical Alloying
12.23c), to form Sm 2Fe 17N x phase [40]. The amount of residual -Fe
depends on the Sm content, and its optimum amount has been found
to be 12.5 at.%. The grain size after milling was typically 1030 nm
and increased to 50 nm after annealing. The magnetic domain size was
found to be 300 nm. Oxygen was found to enhance precipitation of
-Fe and a maximum of 9000 ppm is suggested [41]. Low oxygen is
found to suppress precipitation of -Fe which deteriorates the magnetic
properties. This new phase, Sm 2Fe 17N x, has coercivities up to 2400 kA/
m and uniaxial anisotropy. The phase has also improved the Curie
temperature T c (470C compared to 320C), the saturation magnetization
M s, (1.6 T compared to 1.54 T) and the anisotropy field H A (14 T com-
pared to 8 T) as compared to NdFeB magnets. Therefore, magnets
of SmFeN should have the potential to compare favourably with well-
established NdFeB magnets, if the high anisotropy field can be ex-
plained to create a high coercivity and a processing route can be es-
tablished. An isotropic resin-bonded (Sm 12.5Fe 87.5) 1xN x magnet shows
remanence M r = 0.71, which is the same for mechanically alloyed
Nd 16 Fe 76 B 8 . The temperature dependence of the coercivity and
remanence of this phase is much smaller than that for Nd-Fe-B. At
150C, the coercivity is still 1450 kA/m. The major drawback of this
new hard magnetic material is its limited stability at elevated temperature.
It has been found that 2:17 nitride is stable up to 650C, thereafter
it decomposes into samarium nitride and -Fe. This instability of the
2:17 nitride prevents the application of sintering technique used for Nd-
FeB or SmCo for the production of anisotropic magnets.
SmCo magnets
The MA process has also been used to produce conventional SmCo 5
and Sm 2 Co 17 magnetic materials. To produce these materials, MA of
128
Applications
Sm 16.7 Co 83.3 and Sm 10.5Co 89.5 powders, corresponding to 1:5 and 2:17
stoichiometry, respectively, is performed in a planetary ball mill under
an argon atmosphere [42]. The MA in the case of Sm 16.7Co 83.3 pro-
duces an amorphous phase possibly with traces of free Co, and for
Sm 10.5 Co 89.5 a two-phase microstructure of an amorphous matrix and
Co solid solution. The two-phase region in the case of Sm 10.5Co 89.5
results from a metastable equilibrium between the amorphous phase and
the Co solid solution. The crystallization of amorphous state to the 1:5
structure, SmCo 5 , occurs at about 550C for both alloys, which has
a high coercivity of about 2200 kA/m and a remanence of 0.52T. The
degradation of the remanence M r results at a high annealing temperature
due to Sm losses during annealing accompanied by the precipitation of
Co. The coercivity of Sm 2Co 17 is found to be low due to the lower
anisotropy field of 2:17 phase. To increase the coercivity, the MA pow-
der is annealed at 600C during which Sm 2Co 17 transforms to SmCo 5.
The order of grain size in the annealed material is found to be less
than 200 nm and the single domain particle size is of about 1 m. As
the coercivity strongly depends on grain size when it is less than the
single domain particle size, MA provides hard magnetic powder at low
temperatures without any sophisticated annealing programme. Conven-
tionally, the preparation of SmCo5 magnets is based on high-temperature
heat treatment above 800C followed by a rapid cooling to room tem-
perature. However, a suitable technique for consolidation of these MA
powders is to be evolved. For MA soft magnetic material refer to
Section 12.13.
obtained along with the other phases like Ti5Si 3 and TiN exhibiting de-
composition of the ceramic phase [44]. Using DC plasma spraying, a
Cu/Ti/C MA powder coating of uniformly dispersed TiC in a copper
matrix has been obtained [45]. These coatings have slightly higher hard-
ness due to homogeneously dispersed structures.
Thus, the proper selection of the ceramic phase both from the re-
activity and wettability point of view is essential in order to have the
uniform desired coating [44].
12.10 SUPERPLASTICITY
The MA 6000 oxide dispersion-strengthened, nickel-base superalloy has
been found to possess superplasticity at 1000C [46]. In the hot-
extruded and hot-rolled condition, i.e. prior to zone annealing, the ma-
terial has a grain size of 0.26 m and a dislocation density of 310 9
cm2. It can be believed that at least some of the is evenly distributed
as very fine (30 nm) particles (Fig.12.2, a similar microstructure, may
be referred). The alloy exhibits superplasticity, having a maximum elon-
gation of over 300% and a maximum strain rate sensitivity of 0.47.
Transmission electron microscopy shows no evidence of ordinary
recrystallization and grain growth is slight. However, strain rates of less
than about 1.0 sec 1 alter the initial microstructure and prevent grain
coarsening on subsequent annealing at higher temperatures. Deformation
of the fine grained MA 6000 can be described as a combination of
power law creep and diffusional (Coble) creep [47] (Fig.12.24). With
a threshold stress caused by the presence of particles, it exhibits only
the diffusional creep process. Threshold stresses for dislocation creep
are not observed.
Superplasticity has also been observed in MA IN 90211 aluminium
alloy (Al2%Mg4.4%Cu1.1%C0.8%O) at 447C and a strain rate
of 1.0 sec 1. The alloy has a grain size of 0.5 m. The deformation
at a high strain rate is found to be governed by co-operative grain
boundary sliding, i.e. the movement of grain groups as a unit [48]. Such
a process proceeds as shear of grain groups along two intersecting sys-
tems, having grain boundary surfaces oriented at 45 to 60 with re-
spect to the tensile axis. Superplasticity has also been observed in IN
9021-SiC p composite material (see Section 12.12).
131
Mechanical Alloying
a wide miscibility gap. In addition, the two metals are virtually insoluble
in each other at room temperature. The situation is aggravated further
by a large solidification range of their alloys and a wide density dif-
ference, which greatly increases the kinetics of lead segregation dur-
ing melting and solidification. As a result of all these difficulties, it is
virtually impossible to produce AlPb castings having a uniform dis-
tribution of lead. The MA of elemental aluminium and lead powders
was carried out in an attritor mill for 2 hrs using 1% stearic acid as
a PCA. After degassing at 200C in a vacuum of 10 2 torr, the MA
powder was consolidated by hot pressing at 500C under a pressure
of 60 MPa [50]. During MA, lead particles undergo plastic deforma-
tion without becoming work-hardened as its recrystallization temperature
(3C) is even below the ambient temperature. Thus, the lead parti-
cles are lamellar in shape and remain soft, whilst the aluminium par-
ticles which undergo plastic deformation, are work-hardened and be-
come fragmented. Consequently, these fragments stick to the surface
of the lead lamellae. The consolidated MA alloy has an improved micro-
structure, which is also reflected in an improvement of approximately
250% in the wear resistance of the alloy.
High energy milling of Al, Fe and ferrochromium powders has been
carried out to develop Al8%Fe2% Cr alloy. The reason for using
chromium is that, being a carbide-forming element, it is likely to combine
with the carbon of the PCA, therefore increasing the wear resistance
of the alloy. After degassing, the powder was consolidated by hot press-
ing [51]. The MA material was found to possess a hardness of 162
BHN, which is comparable to several rapidly solidified AlFex-alloys.
The MA alloy has high wear resistance compared to PM and wrought
alloy, and EDAX analysis has shown the presence of Al4 (FeCr) in-
132
Applications
Brasses
The MA technique has also been used to improve the tribological prop-
erties of brasses. The MA of Cu40%Zn blended alloy and solid lu-
bricant WS 2 or graphite (20 and 40 vol.%) has been carried out in a
planetary mill in an argon atmosphere for a period of approximately
525 hrs. It was observed that MA continuously refines the microstruc-
ture with milling hours (Fig.12.26) but in the case of brass WS2, milling
beyond 13 hrs results in the chemical reaction of zinc with WS 2 to form
ZnS due to activation by the energy input of MA. The graphite par-
ticles retain their hexagonal lattice, as confirmed by SAD patterns of
the composite powder, even after milling for a period of 25 hrs. The
MA powder containing WS 2 was consolidated by hot pressing at 620C
at a pressure of 28 MPa, and the powder containing graphite was con-
solidated by cold compaction under a pressure of 700 MPa followed
by sintering at 810C for one hour [52]. The microstructures of these
consolidated MA composites have uniform fine scale distribution of WS2
graphite, as shown in Fig.12.27.
Friction coefficient f (in unlubricated conditions) and wear rate meas-
urements for these MA materials have shown that it could be possi-
ble to decrease the friction coefficient as low as that of the pure solid
lubricants (f = 0.15), indicating that the sliding surface is completely
covered by a thin layer of the solid lubricant.
Fig.12.26 TEM micrograph of Cu40% Zn brass powder particle, milled for 13 hrs.
Note the lattice planes of WS 2 (Ref.52, Metal Powder Report, 48 (11), 20 (1993)
(Elsevier Science)).
Fig.12.27 (right) Micrograph of sintered MA brassgraphite composite (Ref.52).
133
Mechanical Alloying
Steels
Wear resistant steels for tribosystems have also been developed using
MA techniques as discussed in Section 12.2.
12.12 COMPOSITES
Aluminium-base MMCs, having particulate reinforcements such as SiC,
Al2O 3, spinel MgAl2O4 and BN, are being developed. However, a major
problem with these composites is the excessive contamination by iron
from the alloying mill abrasion. Unlike other processing routes for similar
composites, the particulate reinforcement does not enhance significantly
the strength of the matrix that has already been strengthened by the
oxide and carbide dispersions. The matrix (IN 9021) is thermally un-
stable and loses its strength at temperatures above 150C due to dis-
location annihilation, and the coarse SiC particles do not appear to result
in any significant strengthening at high temperatures [53]. Whilst the
particlematrix interface strength is substantial (well over 1000 MPa),
coarse particles of SiC break during fast fracture and the fracture tough-
ness of the composite is low [54]. However, at high temperatures in
the range 425450C and at high strain rates (>0.7 sec 1), the com-
posite material with the IN 9021 matrix shows extended ductility, up
to 300% total elongation [53,55].
Elemental 7010 alloy has been mechanically alloyed with 10% SiC
particles [56]. After degassing at 200C for 2 hrs in vacuum (10 2 torr),
the composite is consolidated by hot pressing at 565C under a pressure
of 110 MPa and hot extrusion at 585C with an 11:1 extrusion ratio.
The consolidated composite was solution treated at 500C and quenched
in ice-water followed by ageing at 175C for 16 hrs. The composite
had a clean SiCPmatrix interface (Fig.12.28) showing minimum interface
reaction. It should be noted that SiC P appears to inhibit the alloying
Fig.12.28 TEM micrograph showing SiCp interface in 7010 matrix (courtesy Prof. P.
Ramakrishnan).
134
Applications
135
Mechanical Alloying
place during the process. Spherical TiC grains with martensite and B2
structures also form during processing, which causes self-propagating
synthesis initiated by the heat of formation of TiC [61].
Surprisingly, the MA process has been applied to polymer-based
composites. The MA of powders of polytetrafluoroethylene and cop-
per or nickel has been carried out. Amorphization through the SSAR
has been achieved in the composite [62].
During MA, in general the softer material tends to form the matrix
and the harder material disperses within it. This tendency, together with
the tendency of mechanically alloyed materials to become harder with
increasing processing, can be utilized to produce complex, multi-level
metal composites. For example, if tungsten, a relatively ductile metal,
is mechanically alloyed with a very fine ZrO2 nickel powder, it results
in the dispersion of ZrO 2 in tungsten. If nickel powder is added and
processed with tungsten ZrO 2 composite powder, a two-level composite
is formed. The hard and brittle tungstenZrO 2 is broken up and dis-
persed in the matrix of more ductile nickel. Contact between the zir-
conium oxide and nickel is minimal in such a composite. Thus, a
number of levels and the relative degree of dispersion of different in-
gredients that can be obtained by this technique are almost limitless.
138
Applications
Fig.12.30 XRD patterns for the: a) 200 hrs milled sample isothermal annealed at 760C
for 30 min, and b) for an as-mixed sample heated continuously up to 1100C without
premilling, i; FeTiO 3 , r; TiO 2, F; Fe, A; -Fe (C) (austenite) (Ref.76).
139
Mechanical Alloying
12.16 OTHERS
Other potential applications include dental prostheses [78] and super-
conductors [7981]. The MA process is being used to produce precursor
powders by INCO alloys based on BaCu and BiSrCaCu alloys for
producing superconducting wires [82]. The use of ceramics for super-
conducting applications is not suitable because they are very brittle.
CONCLUSIONS
At present, MA materials are used not only in the aerospace industry,
but also in the thermal processing industry (furnace fixtures, muffle
tubes, furnace racks, furnace transport rollers, skid rails, mesh belts,
electrical resistance windings, burner nozzles, etc.), the glass industry
(molten glass stirring rods, furnace hearths, tiles, tableware, bushing
used to produce single and multistrand fibres), the energy production
industry (flame stabilisers, fuel cladding in fast neutron breeder reactors,
heat exchanger components in high-temperature gas-cooled reactors,
components for industrial gas turbines), the recreational industry (bi-
cycle frames and fork brakes) and the spray-coating industry. However,
the progress of industrial acceptance of MA materials has been slow,
mainly due to the following reasons:
high cost (though high performance/cost ratio),
reluctance to try new materials,
lack of in-service experience,
non-availability of the necessary product forms (bar, plate, sheet,
wire, tube, forging stock, etc.),
size range; the upper limit usually depends on the casting pro-
duction facilities (at present 500 kg MA 956 plates represent the largest
made), whereas the lower limit (thickness or diameter) may depend on
the working characteristics of the alloy.
The outlook for the growth of MA materials in industrial markets
appears to be a good one. Efforts to increase the availability of the
product forms and lower cost production methods will certainly improve
the market scenario of these materials. The RSMA technique will prob-
ably result in the advent of a new class of extraordinary materials.
References
1. R. Sundaresan and F.H. Froes, J.Metals., 39 (1), 22 (1987).
2. R.C. Benn, L.R. Curwick and G.A.J. Hack, Powder Metal., (4), 191 (1981).
3. K. Murakami et al, Met. Trans., 24A, 1049 (1993).
4. W. Sha and H.K.D.H. Bhadeshiam Met. and Mater. Trans., 25A, 705 (1994).
5. J.J. Fischer and J.H. Weber, Adv. Mat. Proc., (10), 43 (1990).
6. T.S. Chou and H.K.D.H. Bhadeshia, Met. Trans., 24A, (4), 773 (1993).
7. B. Kazimierzak, M. Prigon and D. Coutsouradis, MPR, 45 (10), 699 (1990).
8. E. Kohler, C. Gutsfeld and F. Thummler, Powd. Met. Int., 22 (3), 11 (1990).
140
Applications
141
Mechanical Alloying
46. R. Singer and G.H. Gessinger, In: Proc. Int. Symp. Deformation of Polycrystals,
N. Hansen et al (eds), Riso National Laboratory, Roskilde (1981), p.365.
47. J.K. Gregory, J.C. Gibeling and W.D. Nix, Met. Trans., 16A, 777 (1985).
48. M.C. Zelin, T.R. Bieler and A.K. Mukherjee, Met. Trans., 24A, 1208 (1993).
49. A. Sharma, P.R. Soni and T.V. Rajan, MPR 43 (1), 37 (1988).
50. V. Kumar et al, Unpublished work.
51. G.H. Borhani, A.N. Tiwari and P. Ramakrishnan, Trans. PMAI., 20, 7 (1993).
52. B. Gnther et al, MPR., 48 (11), 20 (1993).
53. Wadsworth, In: High-Strength Powder Metallurgy Aluminium Alloys II, G.J.
Hildeman and M.J. Koczak (eds), TMS, Warrendale, PA (1986), p.137.
54. D.L. Davidson, Met. Trans., 18A, 2115 (1987).
55. R.G. Nieh, Scripta Metall., 18, 1405 (1984).
56. A. Bhaduri et al, Trans. PMAI, 21, 1 (1994).
57. A.P. Miodownik, Met. and Mater. Trans., 27 (10), 3718 (1996).
58. N.R. Rudrappa et al, PMAI Newsletter, 11 (4), 23 (1985).
59. S. Pickering, MPR, 50 (6), 30 (1995).
60. J.W. Pyun and S.I. Kwun, J. Korean Inst. Metals,33 (6), 814 (1995).
61. H.Q. Ye, Acta Mater., 44 (5), 1781 (1996).
62. T. Ishida and S. Tamasu, J. Mater. Sci. Lett., 12 (23), 1851 (1993).
63. S. Surimach, et al, Mat. Sci. and Eng., A134, 1368 (1991).
64. A.G. Escorial et al, Mat. Sci. and Eng., A134, 1394 (1991).
65. G.M. Dougherty, G.J. Shiflet and S.J. Poon, Acta Metall. Mater., 47 (2), 2275
(1994).
66. Z.H. Yan, M. Oehring and R. Bormann, J. Appl. Phys., 72 (6), 2478 (1992).
67. Y. Minamino et al, Met. Trans. JIM, 37 (2), 130 (1996).
68. M.L. Trudean and R. Schultz, Mat. Sci. and Eng., A134, 1361 (1991).
69. J. Kuyama, J. Jpn. Appl. Phys. Lett., 30 (5a), 854 (1991).
70. Y. Tanaka, Mater. Trans. JIM, 37 (3), 265 (1996).
71. J.Y. Huang et al, Acta Metall. and Mater., 45, (1), 113 (1997).
72. H. Kuwano and T. Hamaguchi, J. Jpn. Soc. Powd. Powd. Met., 39 (3), 216 (1992).
73. J. Rawers et al, Met. Mater. Trans., 27A, 3126 (1996).
74. G.B. Schaffer and P.G. McCormick, Mat. Sci. Forum, 88-90, 779 (1990).
75. P.G. McCormick, MPR, 51 (9), 5 (1996).
76. Y. Chen et al, Met. and Mat. Trans., 28A (5), 1115 (1997).
77. J.Y. Huang et al, Acta Metall. and Mater., 44 (5), 1781 (1996).
78. E. Ivanov, Int. symp. Mechanical Alloying, Kyoto, Japan (1991).
79. C. Politis, Physica, 135B, 286 (1985).
80. M. Stubicar et al, Int. symp. mechanical alloying, Kyoto, Japan (1991).
81. V. Mizutani et al, Mat. Sci. Forum, 88-90, 415 (1992).
82. Anonymous, MPR, 43 (10), 698 (1988).
Questions
1. Give a general flow sheet for the production MA aluminium alloys.
2. What are DISPAL alloys? How are they produced?
3. Discuss development of MA high temperature aluminium alloys.
4. Discuss with the help of a flow sheet the production of Al-Li MA
alloys.
5. Name the MA alloys produced at commercial level, discuss their
properties and uses:
(i)Nickel-base superalloys;
142
Applications
(ii)Aluminium-base alloys.
6. Which MA steels are produced on an industrial scale? What are
their applications?
7. What are ODMs? What are their possible uses?
8. Discuss MA steels useful for tribological purposes.
9. How MA techniques can be helpful in developing the following:
(i)High temperature copper alloys;
(ii)Ti-Mg alloys.
10. How MG is helpful in improving ductility of titanium aluminades.
11. Discuss how Ti 5Si 5 can be improved with the help of MG.
12. How MA produces supercorroding materials.
13. Discuss development of MA hydrogen-storage materials.
14. Compare SSE achieved by MA with that by RS.
15. How MA technique is helpful in developing NdFeB magnets.
16. Write a note on MA powders useful for spray coatings.
17. Why MA techniques are helpful in achieving superplastic
behaviour.
18. Give an account of development of tribological materials using MA
techniques.
19. Give an account of development of particulate composites using
MA techniques.
20. What are the advantages of nanocrystalline materials? Discuss one
example where nanocrystallization has been achieved using MA
techniques.
21. Discuss how MA techniques are advantageous in production of tita-
nium powder from ilmenite.
22. List the reasons for the slow acceptance of MA products at indus-
trial level. How can this state of affairs can improved?
143
Mechanical Alloying
144
List of Symbols
Applications
LIST OF SYMBOLS
145
Mechanical Alloying
rd Dispersoid radius
rh Hertz radius
rp Particles radius
S Relative long range parameter
T Absolute temperature
Ta Ambient temperature
Tb Bulk temperature
Tc Curie temperature
T crit Critical temperature
Ti Post-impact temperature
T ig Ignition temperature
Ts Powder surface temperature
T Change in temperature
t Mechanical alloying processing time
tc Time between collisions
t ig Incubation time for combustion to start
V Volume
Vp Volume of powder impacted per collision
x Vial length in Spex mill
xc Critical thickness of amorphous layer
Z Zener Holloman parameter; viscosity of lubricant
f Particle fracture probability
w Particles welding probability
Deformation manifested strains
e& Particles strain rate during collision
c, f Critical strain to fracture
Relative change in bond length
b Rebound yield
Mean free path between collisions
0 Atomic size factor
0 Relative colliding velocity of the ball
r Grain boundary velocity during recrystallization
s Velocity of sound
Density of ball
Stresses experienced by powder during collision
th Threshold stress below which there is negligible creep
u Tensile strength (ultimate)
y Yield strength (0.2 per offset)
Half of impact duration
Angular velocity of horizontal ball mill
c Critical rotational velocity of horizontal ball mill
p Angular velocity of plate in a planetary mill
v Angular velocity of vial with respect to plate in a planetary mill
146
Index
Applications
Index
A CALPHAD method 72
acids 31 carbide dispersion 65
AgCu 27, 125 chemical reactions 51, 133
AgFe 27 chromium carbides 102
Al(7.515)% Ti 117 (CoFe)75Si15B10 32
Al2% Li4.4% Mg 115 closed packed array 12
Al2%Mg4.4%Cu1.1%C0.8%O 130 Coble creep 130
Al3% Li2.2% Mg 114 consolidation 58
Al4% Mg1.1% C0.8% 112 co-operative grain boundary sliding
Al4.5%Cugraphite 131 (CGBS) 128
Al8% Fe4% Ce 118 critical reaction temperature 52
Al8%Fe2% Cr 132 cryomilling 26
AlAl4C3 composite powder 25 crystal-to-amorphous transformation 47
AlFe 27 CuAg 29
AlFeCe 28, 116, 118 CuAl 42
AlFeMn 28 CuAlC 116
AlMg 31 CuC 116
AlTi 116 Cu40%Zn 133
Al2O3Y2O3 105 Cu8% Ti4% B 120
Al50Mg50 33 CuCr 119
aluminium alloys (see dispersion strengthened) CuFe 4, 27
amorphization 21 CuPb 4
amorphization by MA 41 CuV 42
amorphization reaction 48 CuW 42
amorphous solids 136 CuZr 119
Arrhenius equation 58
atomistic models 72, 87 D
attrition 6 dead zone 11
attritor 6 degree of crystallization 20
attritor efficiency 11 differential thermal calorimetry (DTC) 49
attritor mill 3 diffusion welding 92
average flake size 19 diffusivity 53, 45
B Dispal 91, 112, 4, 115
dispersion strengthening (DS) 2
B2 TiAl 26 materials 2
ball mill 2 commercial materials 2
ball rotational velocity 81 superalloys 1, 2, 3, 18, 100, 105
ball to powder ratio 9 iron-base materials 4
ball velocity gradient 11 aluminium alloys 107
balltopowder ratio (BPR) 19, 20 copper 115
brazing 93 dispersoid particle radius 67
brittlebrittle system 37 dispersoid radius 67
bilk temperature rise 85 dispersoid spacing 67
dispersoid strengthening 67
C dispersoid volume fraction 67
CALPHAD 43 double mechanical alloying 28
ductilebrittle system 36
147
Mechanical Alloying
ductileductile system 35 H
dynamic compaction techniques 60
HallPetch relationship 65
E Hall-Petch effect 67
head-on collisions 14
EDAX 128
heat treatment 25, 28, 105, 106, 123
elongated grains 61, 62
Hertz impact theory 84
equation of kneading 27
Hertz radius 14, 84
EPMA 39, 41
Hertzian collision 85
Energy maps 55, 56
HfAl 42
ethylene bis disteramide 31
HfCu 42
exothermic redox reactions 51
HfNi 42
explosive compaction 89
high-angle grain boundaries 49
extrusion ratio 60, 101, 113
high-energy milling mode 15
F high-speed blenders and shakers 7
horizontal mills 81
fatty acids 2 hot isostatic pressing (HIP) 60
Fe13.5%Cr0.3%Mo1%Ti0.3%Y2O3 hot upsetting 104
107 HSLA steels 67
Fe20% Cr4.5% Al0.5% Ti0.5% Y2O3 hydrogen storage materials 4, 123
107 hydrostatic stress 39
Fe2B 22, 23
FeCrAl 109 I
fluid dies 59
ignition surface coating technique 2
fussionfusion 77
ignition temperature 52
forged bonding 94
ilmenite 139
friction coefficient 133
immiscible liquids 4
fracture
immiscible solids 4
dynamic 72
impact energy 20
forging 72,74
impact stress 14
shear 72
impact time 83
free-ball velocity 14
IN 9021-T4 113
forging 60,62
IN 9052 112
IN-905 XL 116
G IncoMAP 111
INCONEL MA 754 103
gas gun launcher 60 incubation period 52
GeSi 37 initial grain diameter 58
glass transition temperature 43 internal oxidation 2
glide 50, 66 International Nickel Company 3
global modelling 73 interparticle necking 37
mechanistic 71 interparticle spacing 2
atomistic 71, 86 iron titanate 139
thermodynamic 71, 87
kinetic 71, 87 K
grind limit 17
ketones 31
grinding balls 9, 23
kerosene 2
grinding media 10
kinetic models 72
148
Index
Applications
L N
laboratory planetary mill 9 nanocrystalline cermets 51
large diameter ball mills 9 (see also horizontal nanocrystalline solids 137
mills) nanocrystalline TiN powder 123
lattice strain 46 nanocrystallization 49
lattice defects 15 Nb3Ge 48
liquid phase sintering 61 NbAl 42
liquid quenching model 41 NbCuSi 42
local modelling 73 NbNi 40, 42
localized melting 61 NbSn 48
long-range order (also see LRO parameter) NbC 110
low-angle grain boundaries 50 Nb40Ni60 40
low density alloys 113 NdFeB magnets 126
low energy milling mode 15 Nd16Fe76B8 128
low pressure plasma spraying (LPPS) 125 NiZr 21
LRO parameter 47 NiZr2 48
Ni20%Cr2%ThO2 40
M Ni32TiNi45Nb55 46
MA 6000 alloy 91, 92, 103, 106 Ni60Nb40 23
MA 754 103, 107 Ni60Ti40 23
MA 758 103 nickel-base superalloys 3, 103
MA 760 103, 104, 107 NiTi2NiNb 46
MA 956 107 NiZr2 22
MA 957 107 NiZr2Ni11Zr9 37
magnesium-base materials 123 O
magnetic materials 126
mean free path between collisions 80 occluding air 5
mechanically activated chemical reactions ordering 56
138 Orowan mechanism 66
mechanical grinding 15 Orowan strengthening 69
mechanical interlocking 58 Orowan stress 70
mechanistic model 85 OrowanAshby expression 67
mechanistic models 72 Ostwald ripening 1
metastable hydrides 25 oxidation 1
Meyers hardness coefficient 68 oxidation resistance 103
MgAl2O4 134 oxide coated balls 10
MgO 112 oxide dispersion microforged material
MgZn 16 (ODM) 109, 110
Mg70Zn30 15 oxide dispersion 65
microcracks 38 oxygen/carbon ratio 32
microhardness 18, 29
mill parameters 19 P
mill speed 19, 21 packing factor 81
mixing technique 2 pair distribution function 86
modelling mechanical alloying 72 particle thermal conductivity 87
modified attritor 11 particle welding 15
morphology 20, 37, 51, 101 pebble mill 3
Mssbauer spectroscopy 19 pipe diffusion 36
MS 100, 101 planetary ball mill 9
149
Mechanical Alloying
150
Index
Applications
X
X-ray technique 39
X-ray line broadening 19 (see also X-ray
technique)
Y
Y2O3 13, 36, 102
Yield strength 64,65,67
Z
ZenerHolloman parameter 62
zone annealing 62,100
ZnFe 42
ZnNi 42
151